Cummins Engine Company v. General Motors Corp.

Full title: CUMMINS ENGINE COMPANY, Inc. v. GENERAL MOTORS CORPORATION, and…

Court: United States District Court, D. Maryland

Date published: May 22, 1969

299 F. Supp. 59 (D. Md. 1969)

Opinion

Civ. No. 15859.

April 9, 1969. As Corrected May 22, 1969.

H. Vernon Eney, Norwood B. Orrick and Venable, Baetjer Howard, Baltimore, Md., Richard Russell Wolfe, Berton Scott Sheppard and Wolfe, Hubbard, Voit Osann, Chicago, Ill., for plaintiff.

John W. Avirett, 2nd, Baltimore, Md., Alfred C. Aurich, Philadelphia, Pa., Arthur C. Raisch, Detroit, Mich., for defendants.

FRANK A. KAUFMAN, District Judge.

This is an action for infringement of certain claims of United States Letters Patent No. 3,110,293, issued November 12, 1963, on an application of Neville M. Reiners filed May 24, 1961 (the Reiners patent). Plaintiff, Cummins Engine Company, Inc. (Cummins) is an Indiana corporation and has its principal place of business at Columbus, Indiana, where it manufactures diesel engines, primarily for use in trucks made by other manufacturers. Cummins is the owner of the entire right, title, and interest in and to the patent in suit and has been since the patent issued. Defendant, General Motors Corporation (General Motors), is a Delaware corporation and has its principal offices in Detroit, Michigan, and has numerous manufacturing, sales and service facilities located throughout the country. General Motors manufactures Toro-Flow engines at its GMC Truck and Coach Division in Pontiac, Michigan, and this division also has a regular and established place of business in Silver Spring, Maryland. Those Toro-Flow engines are the accused engines in this proceeding. Defendant, McCall-Boykin Truck, Inc. (McCall-Boykin), is a Delaware corporation having its principal place of business in Baltimore, Maryland, where it sells and services the accused Toro-Flow engines.

FINDINGS OF FACT¹

1 There are other findings of fact which could have been made from the voluminous record in this case. This Court has not made such additional findings because, in some instances, the evidence presented was inconclusive, and, in others, specific resolution of disputed facts was not required in order to reach the legal conclusions set forth in this opinion, and would in no way have altered any of those conclusions.

A. Diesel and Gasoline Engine Technology.

The Reiners patent in suit concerns the open chamber type of diesel (or compression-ignition) engines. This class of diesel engines is characterized by a mode of combustion involving the direct injection of fuel into the cylinders of the engine. In other respects, however, the class shares certain common operating features with all other classes of diesel engines. On a broader level, the diesel engine itself differs from its predecessor, the conventional gasoline (or spark-ignition) engine, in some respects and is similar or identical in others. In order to assess the significance of the Reiners patent as it applies to one class of diesel engines, it is helpful to examine the similarities and differences that run through gasoline and diesel engine technology.

1. The Four-Stroke Cycle.

Basic to all these engines is a common four-stroke power cycle² operating with the same fundamental elements — a piston, connecting rod, crankshaft, and cylinder head with intake and exhaust valves. Basic also is the kinematics, or mechanical operation, of the engines.

2 The vast majority of internal combustion engines in commercial use operate with a four-stroke cycle. Some internal combustion engines utilize a two-stroke cycle, in which only the power and compression strokes are present. This suit concerns only four-stroke cycle engines.

Figure 1 shows diagramatically the operation of the four-cycles of the internal combustion engine. During the first (or intake) stroke, the piston moves downward and the intake valve opens with the exhaust valve closed. The downward motion of the piston causes air, which is at normal atmosphere pressure, to be sucked into the cylinder. The intake valve is then closed, with the exhaust valve remaining closed, at the start of the second (or compression) stroke, thus trapping the air in the cylinder; the piston moves upward and compresses this trapped air between the top face of the piston and the cylinder head. During the compression stroke, combustible fuel is injected into the cylinder, and at or near the end of this stroke the fuel-air mixture is ignited. With the intake and exhaust valves remaining closed, the heat of combustion raises the temperature and pressure within the cylinder; this correspondingly expands the trapped gases and drives the piston downward, initiating the third (or power) stroke. The downward motion of the piston acts through the connecting rod to rotate the crankshaft of the engine. The energy thus delivered to the crankshaft can be used to drive the equipment coupled to the engine (usually through a clutch and transmission arrangement). All the useful power of the engine is derived from the third stroke of the cycle. Near the end of this stroke, the exhaust valve opens and, as the piston moves upward during the fourth (or exhaust) stroke, the spent gases are forced out of the cylinder through the exhaust port. This clears the cylinder for another intake stroke and thereby completes the four-stroke cycle.

2. Combustion System Designs.

The differences between the gasoline and diesel engines are manifested in the second stroke of the cycle. In a diesel engine, the fuel is ignited by the heat generated from highly compressing the trapped air. Because a diesel engine utilizes the heat of compression to ignite the fuel, a relatively high degree of compression is required. The ratio of the volume of the air before compression (which is equal to the volume of the cylinder and is called the displacement) to the volume of the air at the end of the compression stroke when it is squeezed into a small space (defined as the clearance volume) is, for diesels, usually 16:1 or more. Ignition takes place when, as the piston approaches the top of the compression stroke, a determined amount of fuel is injected under high pressure into the clearance space and is mixed with the highly-compressed, very hot air (see Figures 2, 3). In order

to achieve proper combustion, the injection of the fuel, the mixing of the fuel with the compressed hot air and the ignition must all occur within a very short period of time, on the order of about 1/500th second.

On the other hand, in a gasoline engine, the fuel and air are premixed in a carburetor and intake manifold and the combustible mixture is then drawn into the cylinder during the intake stroke. When the piston is at the top of the compression stroke, the air-fuel mixture is ignited by an electrical spark. Combustion thus does not depend exclusively on compression, and this allows the gasoline engine to operate with significantly lower compression ratios, usually of the order of 8:1 or about half the compression ratio of a diesel engine (see Figures 2 and 3).

The higher compression ratio in the diesel engine of course means that there is a higher expansion ratio of the gases during the power stroke. More power per cycle can thus be generated, all other things being equal, and this adds to the efficiency of the diesel engine.³ At the same time, the higher cylinder pressures in a diesel engine mean that the principal engine parts including the cylinder head, cylinder blocks, pistons, connecting rods, crankshaft and associated bearings must be stronger and heavier than the smaller parts in a gasoline engine; and a design problem exists to minimize the extra weight and cost of these parts. Differences in cylinder pressures are, however, differences of degree only since the kinematics of the diesel and gasoline engines are basically the same. Knowledge of developments in gasoline engine design to accommodate high pressures is therefore adaptable to diesel engine design, and it is undisputed that an engineer skilled in the art would have no difficulty in designing the engine parts to withstand whatever additional pressures are encountered.

3 The higher expansion ratio of the diesel engine is the principal reason that it has a greater theoretical efficiency than the gasoline engine. Two other factors contribute, though in a lesser degree, to this result: diesel fuel has a somewhat higher thermal energy content than gasoline; and carburetor losses in premixing gasoline and air outside the cylinder are not present in those diesel engines in which the fuel is injected directly into the cylinder. In this case, we are dealing with diesel engines of that type.

Differences in the modes of combustion, however, create an additional problem for the diesel engine designer that cannot be solved by reference to the gasoline engine art. Because in a diesel engine the fuel must be injected, thoroughly mixed with the air and ignited within a very small period of time and in a small space, the clearance volume of the cylinder must be accurately designed and matched with an appropriate injector spray. Tolerances of error in both design and machining must be minimized, and as the compression ratio is increased for a given displacement this problem is accentuated.

Some diesel designers have attempted to alleviate this problem by using a divided combustion chamber⁴ into which all or part of the fuel is injected or directed to mix with the air and to burn partially before being expelled through a passageway into the engine cylinder where it acts as a catalyst for combustion. Though proper combustion is thereby facilitated, pumping losses and heat losses are increased, so engine efficiency and fuel economy are sacrificed to some extent.

4 Several names for divided combustion chambers, depending on their particular designs and operations, are air cells, energy cells, turbulence cells and precombustion chambers.

The diesel engines of the patent are characterized as being of the open chamber type. In such engines, the fuel is injected directly into the cylinder as the piston nears the top of the compression stroke. In order to achieve proper combustion, the fuel must be atomized into fine droplets, the spray must be dispersed and the droplets must penetrate into the clearance volume. A common combustion chamber design, in widespread use long before the Reiners patent, is based on a Mexican Hat, or Hesselman, chamber. This arrangement is illustrated in Figure 4. The cone-shaped

combustion chamber is, in effect, grafted out of the top of the piston, with the sides of the piston reaching, at the top of the compression and exhaust strokes, practically to the cylinder head. During the intake stroke, the air entering the cylinder is given a rotational movement around the cylinder axis due to the shape and location of the intake ports and manifolding. This effect is called "swirl," and its particular properties influence the choice of the angle and position of the cone in the Mexican Hat. As the piston approaches top dead center on the compression stroke, the rate of swirl accelerates; at the same time, the peripheral (top) portion of the piston forces air radially inward. This latter effect is called "squish". Also to be considered in designing this combustion system is the type of injector to be used: the pressure of the fuel injection, along with the location and sizes of the orifices in the injector, must be chosen and matched with the chamber design. It can be seen that there are many interdependent variables that have to be considered, both by themselves and in relation to each other, before an optimum design can be achieved. The state of the art at present (and, a fortiori, the state of the prior art) is such that there is no exact mathematical formulation that can yield the best working design; instead, there must be resorted to extensive cut-and-try experimentation. Also, some sophisticated minor design improvements — e.g., recessing the valves to decrease the amount of metal in contact with the burning gasses — may be attempted by further experimentation. It should be emphasized that all of this was thoroughly described in journals of the trade and in common use in the prior art long before the Reiners patent was filed.⁵

5 See, e.g., Judge, High Speed Diesel Engines 86-87 (4th Ed. 1941), admitted into evidence as DX 118.

3. The Stroke-to-Bore Ratio.

Commercial pressures to develop engines with more power in a smaller package induced diesel and gasoline engine designers to adopt a variety of techniques to achieve more compact machines. These methods included utilizing better lightweight materials, streamlining, and using the "V" cylinder design.⁶ Another technique was to reduce the ratio of the length of the stroke to the diameter of the cylinder for a given displacement. This relationship is called the stroke-to-bore ratio (s/b) and an engine is said to be "undersquare" when the bore is less than the stroke (i.e., s/b 1) and "oversquare" when the bore is greater than the stroke (i.e., s/b 1). The trend over the years in gasoline engines was to lower and lower stroke-to-bore ratios and had dropped as low as 0.75 before the patented engine was designed. For example, the prior art Chevrolet 348 gasoline engine, which was known to Reiners and other engineers at Cummins prior to the development of the patented engine, had a stroke-to-bore ratio of 0.79. As Cummins' expert witness, Professor Shreeve, admitted, diesel engine design is aided by the experiences in gasoline engine improvements since the latter shows what can be done from a mechanical point of view; and diesel design has followed generally after the trend established in gasoline design, and, particularly, in a gradual downward trend in stroke-to-bore ratios. When Reiners began developing the patented engine, there had already been some movement in diesel engine design towards the use of lower stroke-to-bore ratios.⁷ Reiners knew of Cummins' prior art engines having stroke-to-bore ratios between 1.2 to 0.909 and also of competitors' prior art engines having stroke-to-bore ratios in the range of 1.0 to 0.96. The Cummins engines included the prior art Vera engine, which had been designed by Reiners himself but for which no patent was sought or obtained. The Vera had the lowest stroke-to-bore ratio (0.909) that Reiners had yet tried.⁸ As Professor Shreeve pointed out, proper design practice in choosing a new stroke-to-bore ratio is to examine the prior art and experiment at a figure slightly below that lowest in operation.⁹ And the chief executive and two engineers of Cummins recommended, on the basis of the operation of the Vera, that experiments be conducted to achieve a still lower workable stroke-to-bore ratio.¹⁰

6 The importance of these factors is discussed infra in the discussion of the prior art Vera engine.

7 Tr. 661-62, 665, 2140-41. Cf. PX 9, p. 579.

8 All the diesel engines referred to in the text are of the open-chamber type. In addition, diesels with precombustion chambers with stroke-to-bore ratios of less than 0.9 had been built and tested experimentally as early as 1956. Tr. 64-68, PX 9.

9 Tr. 660.

10 See DX 8, DX 10, Tr. 408-17, 446-48, 518-20, 530 (Reiners).

The advantages in compactness yielded from using lower stroke-to-bore ratios for a given displacement are the same for both gasoline and diesel engines. First, by holding the piston speed constant and shortening the stroke, the number of revolutions per minute (rpm) of the crankshaft is increased in a corresponding percentage.¹¹ This increases the horsepower per cylinder.¹² Second, a reduction in the stroke enables the length of the connecting rod and the cylinder block deck height to be reduced. Thus, a savings in weight should also follow from using a lower stroke-to-bore ratio. Third, for a given horsepower, a higher rpm of the crankshaft means transmission is lower, thus permitting that the torque, or twisting force, on the the use of a smaller and lighter transmission. The upshot of all this is that, for both gasoline and diesel engines the use of lower stroke-to-bore ratios, all other things being equal, should yield a lower engine specific weight¹³ at a higher rated speed. Unfortunately for diesel designers, however, all other things are not equal — to wit, the relative difficulty in achieving a good diesel combustion system. And this problem is aggravated as the stroke-to-bore ratio is lowered. In designing a short-stroke diesel engine everything could be borrowed from the evolving gasoline engine technology except the solution to the combustion problem. For this reason, the general consensus of diesel designers, prior to the Reiners patent, was that optimum stroke-to-bore ratios for diesel engines were in the range of 1.0 to 1.4, and the most commonly used were in the range of 1.2 to 1.4.¹⁴

11 See Tr. 78-80, 960-64. Reiners says that he found from experience that 2100 ft/min. was the maximum acceptable piston speed. General Motors' expert, Mr. Rosen, agrees.

12 The horsepower per cyclinder of a diesel engine is a function of three design factors: the brake mean effective cylinder pressure (P), the displacement of the cylinder (V, which is equal to the stroke times the area of the cylinder face), and the rpm of the crankshaft. If P and V are held constant and the rpm is increased a certain percentage, then the horsepower per cylinder will increase by the same percentage.

13 The specific weight of an engine is defined as the ratio of the total weight of the engine in pounds to its rated output in horsepower. For example, a 200 hp engine that weighs 2000 lb. would have a specific weight of 10 lb./hp. The rated speed of an engine is the rpm turned by the crankshaft in developing the power output at which the engine is rated.

14 See PX 7B, PX 91, Tr. 1287-89 (Rosen). Cf. PX 7A, C, D.

B. The Patented Engines.

1. Development.

During the summer and fall of 1959, Reiners directed several members of the Cummins research staff to conduct independent theoretical analyses of the effects on fuel efficiency of decreasing the stroke-to-bore ratio. Four factors caused by a lower stroke-to-bore ratio influence fuel efficiency. By increasing the bore relative to the displacement, larger intake valves could be placed in the cylinder heads, thus giving freer air "breathing" on the intake stroke. Improvement in pumping losses and combustion efficiency could thereby be anticipated. On the other hand, increasing the crankshaft rpm meant that losses from increased mechanical friction would occur; and losses from lower thermal efficiency could also be predicted due to the greater percentage of metal surface around the combustion chamber. Reiners balanced these improvements and losses over a range of stroke-to-bore ratios and found the maximum improvement over the in-line engine with a stroke-to-bore ratio of 1.2 to be at 0.95.¹⁵ Reiners decided to attempt to build an engine at a stroke-to-bore ratio of 0.75 because he felt that the advantages in weight and economy that could be gained would offset the theoretical decline in combustion efficiency.¹⁶

15 Tr. 135-45, PX 18, 19.

16 Tr. 146.

Reiners then began to experiment in order to find a working combustion system design. After much cut-and-try testing, he settled on a system practically identical to the one he had used in the Vera, which, in turn, had been well-known long before the Vera itself was designed. Indeed, Reiners admits that the only real difference between the Vera and the Vim (the Vim is one of four engines constructed under the patent here in suit, the others being the Vine, the Val and the Vale) combustion systems was the dimensions¹⁷ (see Figure 5). The experts for both sides,

17 Tr. 173, 505-09.

Shreeve and Rosen,¹⁸ agree that the combustion system as a whole used in the Vim was common knowledge prior to the Vim and that there was not one single element in that system not well-known to the trade long before the Vim was designed.¹⁹ The new engine consisted, therefore, of a design using a stroke-to-bore ratio (0.75) already in use in gasoline engines, with a combustion chamber commonly used in open-chamber diesel engines.

18 This Court was fortunate in receiving the testimony of the independent expert of each party — Mr. Rosen (for General Motors) and Professor Shreeve (for Cummins). Both men are remarkably qualified and articulate, and their candid and lucid testimony was invaluable to this Court's understanding of the engineering problems involved in this case. Mr. Rosen has been working on the research and development of diesel engines for over fifty years. During that period, he has delivered over two hundred papers and obtained thirty patents in the field. His experience in diesel engine design encompasses work for the Government as well as private industry and universities. Professor Shreeve is currently the chairman of the Department of Mechanical Engineering at the University of Maryland. He has been a mechanical engineer for over thirty years and has taught at the Pratt Institute, George Washington University and the University of Maryland.
In addition to the independent experts, Rosen and Shreeve, employees of each party testified as experts — Mr. Reiners for Cummins and Mr. G. Flynn for General Motors. Both of these gentlemen have a great deal of experience in the field of diesel engine design and development and their testimony was most helpful to the Court. Finally, skilled counsel patiently labored to reduce complicated questions to the fullest extent possible and to translate the language of a technical field into more ordinary English.

19 Tr. 667-75, 998-99.

The first Vine engine was tested in the Cummins research laboratory before the principal officers and directors of Cummins on May 24, 1960.

2. The Patent Applications.

On May 24, 1961, Reiners filed his first application for a patent. In that application, he stated that the objects of his new engine were:²⁰

* * * to provide a novel compression-ignition engine having a more compact structure for a specific power output, than conventional engines as heretofore constructed.

* * * to provide a novel compression-ignition engine having increased specific power output over that of conventional compression-ignition engines as heretofore constructed.

* * * to provide a novel compression-ignition engine having increased engine efficiency and durability over such conventional compression-ignition engines.

* * * to provide a novel compression-ignition engine which has less length and less height for a given horsepower rating than conventional compression-ignition engines as heretofore constructed.

* * * to provide a novel compression-ignition engine for automotive vehicle use, which permits the use of lighter frame construction for supporting the engine in the vehicle and a lighter transmission, than has heretofore been used.

20 DX 73, p. 1.

Reiners then acknowledged that short-stroke gasoline engines have been commercially built, but that commercial activity in the diesel field had been unsatisfactory "because of the basic problem of obtaining good combustion efficiency in the short stroke diesel."²¹ The solution given in the text is to use a Mexican Hat arrangement with: a minimum practical limit for the cylinder bore about four inches, the injector located approximately on the cylinder axis, the valves recessed into the cylinder head, the "squish" effect taking place, and the use of proportionally larger valves. Reiners cautioned that careful design was necessary.²² Included in the application was a drawing showing the cross section of the combustion chamber.²³ In addition, the text pointed out that:

21 Id. p. 2.

22 Id. pp. 3-5, 6-7.

23 Id. p. 17 Figure 4. See also id. p. 15, Figure 1.

* * * the length of the connecting rods * * * may be reduced. In any engine, the connecting rods are of course made as short as possible without creating an excessive side thrust on the walls. With reduced length of the connecting rods, the cylinder block and hence the overall size of the engine may be reduced. Consequently, the ratio of the length of the piston rods to the bore of the cylinders is reduced, and such ratio in an engine embodying the invention is less than 1.75.

The reduction of the length of the connecting rods also provides a reduction in the distance from the crankshaft axis to the outer surface of the cylinder block, that is, the surface of the block at the outer or upper end of each cylinder. * * * The reduction in such distance [called the "deck height"] of course provides a more compact engine. With the reduction in such distance, the ratio of such distance to the bore of the cylinders is reduced, and in the engine of the invention this ratio is less than 2.8.

These values did not purport to contribute to the efficiency of the combustion system in any way. They were stated as significant only with respect to the compactness of the engine. An illustrative drawing was included in the patent application (see Figure 6).

On the basis of this discussion, Reiners stated a claim in his patent (Claim 1) in which the only parameter was a stroke-to-bore ratio of less than 0.9.²⁴ But aside from stating that the minimum practical limit for the cylinder bore is about four inches, the patent application gave no dimensions whatever. It was completely silent as to any specific diameter of the cylinder bore, length of the stroke, stroke-to-bore ratio, length of the connecting rod, connecting-rod-to-bore ratio, deck height and deck-height-to-bore ratio. And more importantly, in light of the statement that the key problem was obtaining good combustion efficiently, the application omitted quantitative values or dimensions as to the combustion chamber (except that the bore should be 4" or greater), the fuel injector pressure, the number and the dimensions of fuel injector spray holes. Swirl and swish are recognized as existing but are not analyzed with respect to the engines described. Some ratios can be scaled off the accompanying drawings, but the prospective designer is left to experiment to fit the fuel injector and combustion chamber. Finally, there is no discussion in the patent of heat rejection, engine cooling or radiator capacity or size.

24 Fifteen proposed claims were included in Reiners' first application. All those claims were made with respect to the open-chamber type diesel engine with a plurality of cylinders, a piston in each cylinder, a crankshaft, and a connecting rod connecting each piston with the crankshaft. Claim No. 1 covered all such engines with a stroke-to-bore ratio less than 0.9. Claim No. 14 covered all such engines with a connecting-rod-to-bore ratio of less than 1.75 (though not necessarily with s/b less than 0.9), and Claim No. 15 covered all such engines with a deck-height-to-bore ratio of less than 2.8 (though, again, not necessarily with s/b less than 0.9). Claims No 2 through 13 covered engines described by Claim No. 1 and set forth certain design features. None of the claims specify a minimum bore dimensions of 4". Id. pp. 10-13.

Reiners' claim to an open-chamber diesel engine with a stroke-to-bore ratio of less than 0.9 was rejected as unpatentable.²⁵ The Examiner said, inter alia: ²⁶

25 Id. pp. 18-20.

26 Id. p. 19.

* * * Applicant's attention is directed particularly to the patent of Nichols. In column 2, lines 35-46, Nichols recognizes the basic problem which applicant is apparently attempting to solve — that of reducing the stroke-to-bore ratio without the resultant impinging of the fuel sprays on the cylinder head and piston surfaces. Nichols' solutions to the problem, as is evidenced in his drawings and discussion, are the same as applicant's solutions, yet Nichols recites a stroke-to-bore ratio greater than unity. With the same structural and theoretical solutions as Nichols discloses applicant selects a ratio of less than 0.9 and thereby secures a more compact engine by lowering the height thereof. This does not amount to invention inasmuch as the shortening of the piston stroke would obviously result in lowering the engine height.

Reiners thereupon cancelled all claims then being sought and submitted an amendment to his application.²⁷ The text and drawings were left unchanged, but seven new claims were set forth. In each claim, one or more parameters additional to the stroke-to-bore ratio were added — e.g., the connecting-rod-to-bore ration, deck-height-to-bore ratio, and minimum diameter of the bore. Claim 2 includes all the elements recited in these new claims which are in suit:²⁸

27 Id. p. 21.

28 Claims 2, 3, 4 and 7 of the Reiners patent are in suit. Claim 3 differs from Claim 2 in omitting reference to a minimum bore diameter and in not calling for an injector or valves. Claim 4 is substantially the same as Claim 2, except that it does not specify an injector or valves. Claim 7 differs from Claim 2 in omitting the connecting-rod-to-bore ratio and the deck-height-to-bore ratio. For convenience, the four claims of the patent which are in suit will sometimes hereinafter be referred to simply as "the claims."

A compression-ignition engine comprising

a plurality of cylinders of the open chamber type,

a piston in each cylinder,

a crankshaft,

and a connecting rod connecting each piston with the crankshaft

the ratio of the stroke of each of said pistons to the bore of said cylinders being less than 0.9,

and the bore of said cylinders being not less than 4 inches

the ratio of the length of said connecting rods to the bore of said cylinders being less than 1.75

and the ratio of the distance from the crankshaft axis to the remote end of said cylinders to the bore of said cylinders being less than 2.8

each of said cylinders having an injector located substantially on the axis of the cylinder,

and valves at said remote end of the cylinder,

said valves having openings extending from adjacent the injector to substantially the circumference of the cylinder,

the openings of said valves for each cylinder thereby being relatively large for the cylinder displacement of the cylinder having the aforesaid stroke-to-bore ratio,

whereby said engine has a compact structure and a high specific power output.

The introduction of these new claims satisfied the Examiner, and the patent was issued on November 12, 1963. The prior art references cited by the Examiner did not include the prior art Cummins Vera or the Ford 172D.²⁹

29 Nor did they include other prior art discussed in this opinion infra at 80-82. See DX 73, p. 30.

With respect to the numerical parameters in the claims, the claims of the patent are stated in a form similar to that of other patents regularly issued by the Patent Office in the internal combustion art as well as other fields.³⁰

30 Tr. 2331-38.

C. The Prior Art.

Cummins does not contend that it is the combustion systems used in the Vim/Vine that made those engines unique and patentable. Indeed, the claims in the patent are broad enough to cover shortstroke diesels of the open-chamber type with a plurality of cylinders regardless of the combustion design that is utilized. The combustion design illustrated in the patent drawings is relevant only in that it shows a designer enough for him to be able to construct a commercially operable engine. Inasmuch as the drawings do not reveal anything novel in that combustion design, the patent is, in effect, informing prospective designers that adequate combustion can be achieved by experimenting on the Mexican Hat arrangement. For this purpose, dimensions need not be shown, nor must the drawings be to scale. The patent specifications taken together with the accompanying drawings provide sufficient data to enable a designer skilled in the art to achieve acceptable combustion, but only after considerable cut-and-try experimentation based on the arrangement illustrated and described in the patent.³¹

31 Compare Tr. 651 (Shreeve) with Tr. 1444-46 (Rosen).

Cummins' argument is that the uniqueness of the engines described by the patent lies in the dimensional parameters that are specified — to wit, a stroke-to-bore ratio less than 0.9, a connecting-rod-to-bore ratio less than 1.75, and a deck-height-to-bore ratio less than 2.8. Cummins urges here, as it did to the Patent Examiner after amending its application, that it is the cooperation of these three reduced ratios, taken together, that defines the novel diesel engine of the patent. By utilizing these three reductions together, Cummins argues, a small and compact engine of high specific power output may be built; the same result will not follow from reducing only the stroke-to-bore ratio. It is therefore necessary to assess the significance of these ratios, from both an engineering and legal point of view.

First, as all the experts agree and as the patent itself recognizes, in any engine the connecting rods are made as short as possible.³² The length of the connecting rod is determined entirely by mechanical features.³³ The main factor in determining this length is the piston stroke; in the patented engine, for example, the stroke accounts for about 83 percent of the connecting rod length.³⁴ It therefore follows that a reduction in the stroke of the engine permits an approximately corresponding reduction in the length of the connecting rod.³⁵ And, as Reiners testified, this means that the connecting-rod-to-bore ratio is thereby decreased:³⁶

So, therefore, the lower the stroke-to-bore ratio, the lower the stroke is and, therefore, the lower the connecting rod length can be.

32 Tr. 656 (Shreeve), Tr. 94-95 (Reiners), Tr. 987-88 (Rosen), PX 6, col. 4, line 3.

33 Tr. 94 (Reiners).

34 Tr. 95 (Reiners).

35 Tr. 95, 124 (Reiners).

36 Tr. 95 (Reiners).

* * * * * *

Now, the connecting rod length is related to the bore directly because of the fact that the stroke is directly related to the bore in the stroke-to-bore ratio.

And since the stroke is responsible for the major portion of the connecting rod length, then the connecting rod length is related to the bore through these relationships.

And we have established that this ratio is 1.75 or less in the patented engine.

The reduction of the deck height (and the deck-height-to-bore ratio) also follows as a matter of standard design practice.³⁷ This too is spelled out by Reniers:³⁸

Now, when the piston is at top dead center, the deck height is established by first of all half of the stroke from the center line of the main bearing to the center line of the crank pin. It is made up of the total length of the connecting rod from the center line of the crank pin to the center line of the piston pin, and the remaining distance is from the center line of the piston to the top of the crown of the piston, plus the running clearance that is necessary to prevent interference of the piston with the cylinder head.

So the connecting rod is a major factor in determining this height, as well as the half of the stroke.

Now, since the stroke is directly related to the bore, the connecting rod is directly related to the bore, it follows that the deck height can, through these interconnections, be related to the bore, and we have established for the patented engine that this ratio of the deck height to the bore should be less than 2.8, to achieve the advantages we seek.

37 Tr. 96-97 (Reiners), Tr. 989-90 (Rosen), Tr. 656 (Shreeve).

38 Tr. 96-97.

In sum, therefore, there is a direct relationship between these ratios.³⁹ A reduction in the stroke of the engine enables a designer to reduce both the length of the connecting rod and the deck height by standard practice. The other factors which limit these parameters — e.g., the counterweight radius on the crankshaft, the length of the piston skirt below the center of the piston pin, the side thrust of the piston, the number of cylinders in the engine, etc. — are common to all diesel and gasoline engines and are considered as a matter of usual practice in streamlining the engine. The Reiners patent does not add anything new in these regards.

39 Tr. 126 (Reiners).

Second, there is general agreement among the experts that there is no abrupt improvement in the operation of the type of engine involved here with the three ratios per se specified by Reiners, as contrasted with an engine with ratios slightly above those values.⁴⁰ In other words, none of these ratios themselves, either separately or in cooperation, establish a critical, threshold point where the engine exhibits distinctive qualities. Small dimensional changes give rise to proportionate, rather than abrupt, changes in engine characteristics.⁴¹ Third, the use of the stroke-to-bore, connecting-rod-to-bore and deck-height-to-bore ratios specified in the Reiners patent was pioneered in the automobile industry in the evolution of compact high-powered gasoline engines.⁴² For example, the prior art Chevrolet 348 gasoline engine had a stroke-to-bore ratio of 0.79, a connecting-rod-to-bore ratio of 1.49, and a deck-height-to-bore ratio of 2.5.⁴³ Cummins argues that this is legally irrelevant to developments in open-chamber diesel engines. But, as this Court has pointed out above, with almost tiresome reiteration, the differences between the gasoline and diesel engine technology lie in their contrasting modes of combustion, not in their mechanical operations. In considering how those ratios are used to design a small and compact engine with a high specific power output, the gasoline experience is part of the relevant prior art.

40 Tr. 294-305 (Reiners), Tr. 1029 (Rosen), Tr. 2149-51 (Shreeve).

41 See PX 18A (smooth curve). The text of the Reiners patent specifies that "the minimum practical limit for the cylinder [diameter]" should be "about four inches." The patent claims an engine with "the bore of said cylinders being not less than 4 inches." Reiners testified that four inches was a critical limit for the bore in that good combustion could not otherwise be obtained; "about four inches" means "exactly four inches." (Tr. 478). Reiners stated that he concluded this from his experience but did not point to any experiments he had done; nor did he give any other objective evidence for this conclusion: nor is there anything in the theoretical analyses underlying the patent and the development of the patented engines which indicates that Reiners held this conclusion prior to his patent application. Reiners' conclusory "exactly four inches" testimony does not, in and of itself, or taken in the context of the entire record in this case, establish that four inches is a critical point for the bore of short-stroke open-chamber diesel engines.

42 Tr. 438-44 (Reiners).

43 Ibid., DX 8, p. 6 and chart 1 (p. 1). The bore of the Chevrolet 348 is 4 1/8", which satisfies the 4" minimum set by Reiners in his patent.

Fourth, the terms "connecting-rod-to-bore ratio" and "deck-height-to-bore ratio" appear to have been coined by Reiners. There are no references or mention of these ratios in either the diesel literature prior to the Reiners patent (including the Cummins bulletins) nor the diesel literature following the patent.⁴⁴ Even the publications of the experts relied on by Cummins contain no reference to those ratios. In Reiners' own work in developing the patented engines, the principal analysis is on the effect of lowering the stroke-to-bore ratio, and only scant allusions are made to the connecting-rod-to-bore and deck-height-to-bore ratios.⁴⁵

44 PX 10, 12, 12A-G, 128-31, 148, 158, 180, 187B, 187C, 188; DX 2, 13, 95, 96, 119.

45 In PX 18, which Reiners said was the basis for his decision to build the patented engines, there is no mention at all of connecting-rod-to-bore or deck-height-to-bore ratios.

A parameter that was in common use by diesel engine designers is the L/R ratio, that is, the ratio of the length of the connecting rod to the crank radius or throw. The common practice was to choose an L/R ratio less than or equal to 4.0.⁴⁶ Since the throw is equal to one-half the stroke, one can translate the L/R ratio into the connecting-rod-to-bore ratio for a given stroke-to-bore ratio.⁴⁷ Fifth, comparisons with prior art open-chamber diesel engines are of course most illuminating and most relevant. For, even given the facts that the ratios of the patent had been used in gasoline engines and the combustion system of the patented engines had been commonly used in diesel engines, Cummins could argue that the engines described by the patent were so different from prior art open-chamber diesel engines that this in itself shows that there was nothing obvious in what Reiners did. An examination of the prior art, however, negates this argument.

46 Tr. 987-88 (Rosen), Tr. 2129-30 (Shreeve). Professor Shreeve, Cummins' expert witness, stated that he did not know of any open-chamber diesel engines with an L/R ratio greater than 4. Ibid. Richardo's basic text, referred to by both experts, recommended a figure of 3.8 or less for under-square six-cylinder engines and suggested that it would have to be somewhat greater for short-stroke engines in order to prevent interference between the bottom of the piston and the crankshaft counterweights. See Tr. 2082-84. The Ford 172D and the Vera have L/R ratios of 3.45 and 4.0 respectively. Tr. 1518-19. Most prior art Cummins diesel engines have L/R ratios in the range of 3.8 to 4.0. Tr. 1519. The only example of a diesel engine with L/R v4 in the record is the GAI 4-51, which has an L/R ratio of 4.1. But that is a two-stroke cycle engine. Tr. 2182-83, 88.

47 This was illustrated by Flynn in his calculations with respect to the Briling DB-43 engine. The stroke of that engine is 92 mm.; thus the throw, which is one-half the stroke, is 46 mm. Choosing an L/R ratio of 3.8, from Ricardo's text, Flynn computed the length of the connecting-rod by multiplying the throw times the L/R ratio. He obtained a value of 175 mm. Using an L/R ratio of 4.0, he did the same multiplication and found the length of the connecting-rod for that L/R ratio to be 184 mm. The bore of the DB-43 was given by Briling as 108 mm. Having calculated the length of the connecting-rod for each L/R ratio, Flynn then proceeded to obtain the connecting-rod-to-bore ratio for each L/R ratio. The resulting connecting-rod-to-bore ratios are 1.61 and 1.685 for L/R ratios of 3.8 and 4.0 respectively. (Tr. 1515-41). The Reiners patent recites a stroke-to-bore ratio of less than 0.9 and a connecting-rod-to-bore ratio of less than 1.75. Using the same approach used by Flynn, it can be seen that an engine with ratios of 0.899 and 1.7499 would have an L/R ratio of 3.89, which comports with the common design practice. The patented Vim/Vine engines, with a stroke-to-bore ratio of 0.75 and a connecting-rod-to-bore ratio of 1.49, have an L/R ratio of 3.98, also within the 4.0 ceiling used by diesel designers.

(a) The Cummins Vera Engine.

The Vera engine, admittedly part of the prior art,⁴⁸ was Cummins' first over-square engine. Design work was started in the early 1950's and proceeded under Reiners' general supervision.⁴⁹ The theoretical horsepower of the Vera is the same as the patented Vine engine and the only difference in the combustion design (other than the use of recessed valves in the Vine) is a dimensional one.⁵⁰ A comparison of the Vera to the engine claimed in the patent shows that they are identical in all respects except that the Vera has a stroke-to-bore ratio of 0.909 instead of 0.9; a connecting-rod-to-bore ratio of 1.818 instead of less than 1.75; and a deck-height-to-bore ratio of 2.909 instead of less than 2.8.⁵¹ General Motors has strenuously argued that these differences are minute and insignificant. Cummins has tirelessly refused to concede this and has forcefully asserted that a comparison between the actual performance of the Vera and the patented engines demonstrates the importance of the patented ratios. Since it is the claims of a patent that measure the grant to the patentee, the controlling comparison is between the prior art and the claims themselves. The patented engines have stroke-to-bore ratios well below the upper limit of 0.9 set in the patent. This Court believes, however, that a comparison between the performance of the Vera and engines constructed under the Reiners patent (the patented engines), though not controlling, is relevant.

48 Pre-trial Stip. II-B-1, 2, pp. 6, 7.

49 Tr. 248-51 (Reiners).

50 Tr. 284, 461-62, 475-78, 490-91, 505-06 (Reiners).

51 Pre-trial Stip. II-B-1, 2, pp. 6, 7. The bore of the Vera is 5.5", or greater than the 4" minimum specified in the patent. Id. at p. 6.

The Vera engine has a rated speed of 2500 rpm and a specific weight of 8.4 lb/hp.⁵² The patented Vim/Vine engines are rated at 2600 rpm and have specific weights of 8.4 and 7.9 lb/hp.⁵³ In arguing that the patented engines are novel, Cummins has relied to a large extent on the comparison of those Vim/Vine operating figures with "conventional" (undersquare) engines for duty in medium-to-heavy trucks that, allegedly, have specific weights in the range of 12 lbs./hp and rated speeds averaging 2100 rpm.⁵⁴ The predecessor Cummins inline engines in a comparable power range have specific weights in order of 9.8 to 11.0 lbs/hp and are rated at 2100 rpm.⁵⁵ It thus appears that the advantages in higher rpm and lower specific weight over the "conventional" engines are equally true of the Vera and the patented engines. Furthermore, all of the five asserted objects of the patent were attained by the Vera.⁵⁶

52 DX 2, p. 33; Pre-trial Stip. II-B-2, pp. 6-7.

53 DX 2, pp. 24, 31; PX 10, p. 5. See PX 11.

54 See Plaintiff's Proposed Findings of Fact 10; Plaintiff's Brief After Trial 11-17. Cf. PX 128, 129; Tr. 101-08 (Reiners).

55 PX 10, 11.

56 In advertising the Vera, Cummins stressed the same objects and advantages as in its later advertisements of the patented engines. These statements, made by Cummins itself, lend support to the defendants' arguments concerning the similarity of the Vera to the patented engines. While this Court generally does not attach any particular weight to advertising claims of this nature, nevertheless when such claims are made by an inventor with regard to one of his earlier devices, they must be considered in connection with the issue of the validity of a patent for a subsequent and similar device.

Cummins insists that the performance of the Vera was "disappointing" from a commercial point of view because it was too big, too heavy and too expensive for use in on-highway trucks.⁵⁷ The Vera engine weighs 2940 lbs., with outside dimensions of 56 11/16" x 45 21/32" x 46 17/32", whereas the patented Vine V8-265 weighs 2080 lbs. and has dimensions of 43 7/8" x 32" x 40 3/4".⁵⁸ Much of the difference in weight and size between the Vera and the Vim/Vine engines can, however, be traced to factors wholly unrelated to the patented features. One of those factors is the streamlining of the cylinder heads, the cylinder block, the crankshaft, and the wall thickness.⁵⁹ Reiners conceded that were he to now design the Vera, he could make it lighter and smaller by applying these streamlining techniques.⁶⁰

57 Cummins does not, however, dispute that it offered the Vera for on-highway uses. See DX 2, Tr. 261, 264-65 (Reiners). It should be noted that the patent claims are not limited to automotive or on-highway applications.

58 DX 2, pp. 31, 33.

59 Tr. 1135-55 (Rosen); DX 111A, 112A.

60 Tr. 315.

It is true that, apart from this streamlining procedure, the decrease in the ratios of the patent accounted for much of the weight differential between the Vera and the patented Vine. Reiners estimated that at least 550 pounds of the 860 pound differential could be so accounted.⁶¹ But this does not prove that the Vera was "disappointing" because its ratios were all slightly above the maxima specified in the patent. For, the Vine ratios are substantially below these maxima, with a stroke-to-bore ratio of 0.75 (patent maximum — 0.9), connecting-rod-to-bore ratio of 1.49 (patent maximum — 1.75) and deck-height-to-bore ratio of 2.39 (patent maximum — 2.8). Increasing the Vera bore a mere 0.21" (i.e., from 5.5" to 5.71") would bring all the Vera ratios within the claims.⁶² The proper investigation is to determine the advantages of an engine, just within the claims of the patent, to the Vera. This Court finds that those advantages would be very small and would not significantly improve the operation of the engine.⁶³ Indeed, those possible gains would be minute compared to the advantages secured by either streamlining, as was done in the Vine, or by going down to the Vine ratios.

61 Tr. 224-230; PX 183.

62 Tr. 293-304 (Reiners).

63 See Tr. 1028-31 (Rosen); Tr. 305 (Reiners); Tr. 2151 (Shreeve).

Cummins argues that even small gains in lightweightness and compactness are commercially important. This is, of course, true. But it does not follow from that assertion that the design improvements that produced those gains are patentable. The diesel engine has for a long period been subjected to various design improvements which have decreased its specific weight and size. This trend has been notable even in engines where the stroke-to-bore ratio is held above 1.2.⁶⁴ Also, depending on the number of cylinders in the engine, a use of the "V" rather than in-line cylinder arrangement can yield as high as a one-third decrease in specific weight.⁶⁵ In the Vera itself, gains from normal streamlining procedures would far outweigh the advantages of an engine just within the ratios of the patent. This Court rejects, therefore, Cummins' argument that the slight differences between the ratios used in the Vera and those recited in the patent are significant, from either an engineering or legal point of view.

64 PX 7-D; Tr. 379-83 (Reiners).

65 Tr. 1028-29, 1299-302, 1307-09 (Rosen).

(b) The Ford 172D Engine.

The Ford 172D tractor engine is also admitted to be part of the prior art. Both its connecting-rod-to-bore and deck-height-to-bore ratios fall within patent claims. However, this engine has a bore of 3.9" and a stroke-to-bore ratio of 0.92. These parameters are slightly outside the minimum bore of 4" and the maximum stroke-to-bore ratio of 0.9.

For the same reasons as discussed in considering the Vera, this Court finds that the difference between the stroke-to-bore ratio of the Ford 172D and of the patent claims is insignificant in the operation of the engine. Further, the combustion system of the Ford is, like the patented engines and the Vera, based on a Mexican Hat, or Hesselman, design. The chamber used in the Ford is, however, relatively narrower than that used in the Vine and Vera, and the injector pressure is correspondingly lower. (In this respect, the combustion design is more nearly equal to the accused Toro-Flow engines (see Figure 5).) These are two alternatives that a designer can use as a matter of choice.⁶⁶

66 Tr. 947-51 (Rosen).

The Ford engine has a rated power output of 59 hp and a rated speed of 2400 rpm. At rated speed and power, its brake specific fuel consumption is rated at 0.460 lb/hp/hr.⁶⁷ This engine was used as a power source in tractors. Its low rated power output and relatively high fuel consumption have always made it impractical for use in trucks.⁶⁸ In order to produce a power output suitable for truck use, its piston speed and engine rpm would have to be increased well above its rated values. As in any internal combustion engine, poor combustion would result.⁶⁹ This engine operated satisfactorily as a power source for tractors.⁷⁰

67 Pre-trial Stip. II-B-3, p. 8.

68 In two certified tests conducted in 1958, the Ford engine, when mounted in tractors, yielded a power output between 39 and 42 hp and a specific fuel consumption between 0.493 and 0.496 lb/hp/hr. Pre-trial Stip. II-B-3, p. 9. The Ford fuel consumption compares unfavorably with both the Vine (0.39 lb/hp/hr at rated power and speed) and the Vera (0.42 lb/hp/hr at rated power and speed). See DX 2, pp. 31, 33. The specific weight of the Ford (9.8 lb/hp) also compares unfavorably to the Vine (8.4 lb/hp) and Vera (8.4 lb/hp). Ford advertised the 172D with emphasis on its compactness, low specific weight and fuel efficiency ( see DX 54, 55) but this Court gives little probative value to such "puffing." See n. 56 supra. Nevertheless, the fact remains that though the Ford 172D did not match the operating characteristics of the Vine or Vera, it worked satisfactorily for the purpose for which it was designed — powering tractors.

69 Tr. 2284-87 (Reiners). See also Tr. 771 (Shipley); Tr. 813-14 (Freele).

70 Ibid.

In the text of the Reiners patent, "the minimum practical limit for the cylinder [diameter]" is described as "about four inches." The bore of the Ford 172D (3.9") is about four inches.⁷¹

71 Tr. 682, 1005 (Shreeve). See also n. 41, supra.

(c) The Briling Articles.

Between 1955 and 1960, N.R. Briling, a member of the Soviet Academy of Sciences, wrote six articles on the short-stroke diesel engine which are part of the prior art.⁷² Those articles, which were published in Russian language technical journals of the U.S.S.R., describe a series of open-chamber diesel engines with the following parameters:⁷³ Stroke Bore s/b Type

DB43 3.61" 4.25" 0.85 In-line, 4 cyl. DB64 3.61" 4.25" 0.85 In-line, 6 cyl. DB67 4.33" 4.91" 0.88 V, 6 cyl. DKS 4.33" 4.91" 0.88 Opposed DB69 4.33" 4.91" 0.88 V, 6 cyl.

All of the Briling articles were published in England and in the United States. The articles were available to the industry, more than one year before Reiners applied for the patent in suit, in the technical libraries of Harvard University, the Library of Congress, the Pentagon, the British Museum Library, the University of California, and the Battelle Memorial Institute.⁷⁴

74 Tr. 864-78 (Defendants' request for admissions 101-51, which are deemed admitted for failure to respond, Rule 36(a), Fed.R.Civ.P.).

No evidence was introduced (apart from the articles themselves) that any of those engines were in fact built and operated.

The Briling articles give an extensive theoretical analysis of the operation of the short-stroke diesel. Briling traced the trend of gasoline engines to a short-stroke construction with decreases in the length of the connecting rod, the deck height, and specific weight, and remarked that diesel design follows the gasoline experience. Each of the five objects or advantages stated in the Reiners patent is asserted and discussed by Briling. In addition, Briling described with much greater detail than Reiners the combustion problem and the techniques that can be used to overcome it. A mathematical solution is attempted to the difficult problem of heat rejection. Briling ventured here into a highly controversial field.⁷⁵ While his formulae have been cited by other experts as one of the basic approaches in this field, they have been criticized, along with all other general formulae that have purported to explain heat transfer in precise mathematical terms, as inadequate.⁷⁶ One of Briling's conclusions from his calculations is that there is less heat rejection in the short-stroke diesel and that this enables the use of a smaller radiator.⁷⁷ The experts of both Cummins and General Motors agree that this prediction was and is erroneous, because, while the heat rejection per cycle is less in the short-stroke diesel, heat rejection over a period of time is in fact greater.⁷⁸ However, all the other advantages asserted by Briling, and discussed by him in great detail, are correct.

75 Tr. 2161-71 (Shreeve).

76 See DX 144; Tr. 2161-71 (Shreeve).

77 See DX 77-A, p. 1; DX 74A, p. 5; DX 76-A, p. 7.

78 Tr. 2111-15 (Shreeve).

In one of his 1956 articles,⁷⁹ Briling presented the factors that go into the choice of a combustion system, including the parameters of the injector (which Reiners had not discussed at all in his patent). Briling then reported the results of his research in designing a suitable system using variations on a Mexican Hat-type chamber, Briling's reports cover the entire gamut in his research — by varying the stroke-to-bore ratio from 0.65 to 0.93 and by using different fuels.⁸⁰ In this and other articles, Briling describes the construction of those engines in great detail and gives their operating characteristics, including fuel consumption, on test stands and in truck use.⁸¹

79 DX 76A.

80 Tr. 2142-48 (Shreeve).

81 DX 75A, p. 22; DX 77A, table 1, p. 3; DX 79A, table 1, p. 2. Cf. DX 74A, p. 14.

The figures reported by Briling for the tested specific weight and fuel consumption of his engines were not as good as he had predicted theoretically.⁸² However, the reported operating characteristics compare favorably to both the prior art and the patented engines. The DB67, a six-cylinder engine in a V arrangement, was stated to have a specific weight of about 8.9 lbs/hp, which is in the range of the patented engines.⁸³ The specific weights of the DB43 and DB64 engines were poorer (14.6 and 11.6 lb/hp respectively), but those are in-line engines, and the DB43 had only four cylinders.⁸⁴ The reported brake specific fuel consumption of those engines was 0.412 lb/hp/hr at a rated speed of 3000 rpm; this is about the same as the patented Val/Vale engines.⁸⁵ As Professor Shreeve stated, Briling's reported figures indicate that he achieved good fuel economy.⁸⁶

82 Tr. 2096-102 (Shreeve).

83 Tr. 2105-06 (Shreeve).

84 Ibid.; Tr. 1466-72 (G. Flynn).

85 Tr. 2121-24 (Shreeve).

86 Tr. 2133.

The Briling articles give specific values of the stroke, bore and stroke-to-bore ratio of each engine. They do not, however, disclose either the connecting rod length or the deck height. Nor do they state specific values for the ratios of those figures to the engine bore.

The experts Flynn and Shreeve scaled Figures 12 and 13 of a printed copy of the article DX76 in the Library of Congress and obtained values of the connecting-rod-to-bore and deck-height-to-bore ratios for the DB43 engine within the claims of the patent. The accuracy of those drawings has not, however, been established by the defendants. The photocopies of those drawings were shown to be distorted and inaccurate and Flynn's measurements of the copy in the Library of Congress, which were limited to the dimensions necessary to obtain the ratios in the patent, showed only that the second generation copy was distorted with respect to the first.⁸⁷

87 Tr. 1654, 1559-69 (Flynn). See also Tr. 2398-418.

Briling did not state what L/R ratio was used in any of his engines. Using an L/R ratio of 3.8, which Ricardo recommended as the upper limit for six-cylinder undersquare engines, Flynn computed the connecting-rod-to-bore ratio for the DB43 to be 1.61. Using an L/R figure of 4.0, which was the upper limit of diesel engine design practice,⁸⁸ Flynn obtained a connecting-rod-to-bore ratio of 1.685. The computed values of deck-height-to-bore ratios were 2.55 and 2.73 for L/R ratios of 3.8 and 4.0 respectively.⁸⁹ Each of the computed ratios falls within the claims of the patent. Since the stroke and bore of the DB64 and DB43 are identical, the above-computed values are the same for the six-cylinder DB64 as well as the four-cylinder DB43.

88 See supra at 79.

89 Flynn's calculations are found at Tr. 1515-41. Part of those calculations may be found supra at n. 47.

All the elements of the Reiners claims except the connecting-rod-to-bore and deck-height-to-bore ratios are disclosed in the text of the Briling articles. Since, as a matter of accepted diesel engine design practice, the length of the connecting rod and the deck height are made as short as possible, with an L/R ratio of 4.0 or less, a person skilled in the art would have, by employing standard design procedures, built an engine within all the parameters of the Reiners claims from the information disclosed by Briling.

(d) Other Publications (all British).

Four technical articles published between 1955 and 1957 describe a VD8-603 Continental engine. This engine has a V design, with a stroke of 4 1/4", a bore of 4 3/4", and a stroke-to-bore ratio of 0.89. It employs a Lanova combustion chamber, which is a type of divided or precombustion chamber.⁹⁰

90 DX 81-84.

The Thaheld engine was disclosed in three 1946 publications as a short-stroke diesel engine with a type of precombustion chamber called an "air cell." This engine had a bore of 4 7/8", a stroke of 3 7/8" and a stroke-to-bore ration of 0.8. It was reportedly used to power aircraft.⁹¹

91 DX 86-88.

Both the Continental VD8-603 and the Thaheld engines clearly fall outside the claims of the patent inasmuch as they employed a precombustion chamber. However, they are still relevant to show that the prior art diesel engine technology had disclosed engines with a stroke-to-bore ratio of less than 0.9.

The Tilling-Stevens "pancake" diesel engine is disclosed in two articles of 1937 and 1938. This engine is described as a horizontally opposed direct ignition engine of the open-chamber type. The stated values of stroke, bore and connecting rod length yield stroke-to-bore and connecting-rod-to-bore ratios within the claims of the patent. No independent evidence was offered that this engine ever went into operation. Neither plaintiff's nor defendants' witnesses could tell whether the pictures in the articles were of actual engines or simply wooden mock-ups for a trade show.

A 1951 article appearing in "Oil Engine" and written by an author using the pseudonym of "Midlander" analyzes the advantages and disadvantages of reducing the stroke-to-bore ratio.⁹² They are basically the same as given later by Briling and then Reiners. In addition, the author noted that some of the disadvantages would be reduced with improvements in bearing materials; those improvements have taken place since 1951. An illustration in the article, which was included by the author as a theoretical example only, described a short-stroke diesel engine with stroke-to-bore and connecting-rod-to-bore ratios within the claims of the patent.

92 DX 85.

D. The Accused "Toro-Flow" Engines.

1. Comparison with Patent Claims.

The accused Toro-Flow engines are identified as Models D351, D478, DH478, D637 and DH637. The D351, D478 and DH478 models are V-6 engines rated at 130, 150 and 170 hp, respectively; the D637 and DH637 models are V-8 engines of 195 and 220 hp.

The D351 engine has a stroke of 3.58 inches and a bore of 4.56 inches, yielding a stroke-to-bore ratio of 0.79. That engine has a connecting rod length of 7.188 inches and a deck height of 11.595 inches; the connecting-rod-to-bore ratio is thus 1.58 and the deck-height-to-bore ratio is 2.54.

All of the other models have 3.86 inch strokes and 5.125 inch bores, yielding a stroke-to-bore ratio of 0.75. The connecting rod lengths and deck heights are the same as for the D351 engines but because of the larger bore, these models have connecting-rod-to-bore ratios of 1.40 and deck-height-to-bore ratios of 2.26.

The parameters of each of these Toro-Flow models thus fall within the claims of the patent. The injectors are located 0.47 inches off the longitudinal centerline of the cylinders; this is substantially on the axis of each cylinder, as specified in the claims. In addition, every other element of the claims is present in each model of the accused engines. The claims in issue therefore read on the accused engines.

2. Design and Development.

From 1953 through 1960, General Motors worked on the design of five different short-stroke diesel engines. In each case, General Motors attempted to convert an existing gasoline engine into a diesel engine so as to use as many gasoline engine parts as possible from common machine tooling. Two studies were terminated at the paper design stage; in the other three projects, engines were built and tested but never went into production.

Each of those studies involved engines with stroke-to-bore ratios of less than 0.9.⁹³ At least two of the tested engines⁹⁴ had ratios well within the claims of the patent. The engines were deficient, however, in their combustion systems: none of the designs employed yielded a value of fuel consumption that was acceptable on the market.

93 It is interesting that one of the paper studies investigated the "dieselization" of a gasoline engine — the V-12 706 — with a stroke and bore identical to the subsequent patented Val/Vale engines.

94 The Chevrolet 348 and the D-401.

Engineers at General Motors were quite impressed by the operating characteristics of the patented Cummins Vim/Vine engines. A Vim engine was purchased and then studied and tested. Beginning in September of 1961, General Motors directed its efforts to dieselizing the GMC-478 gasoline engine; again a principal objective was to retain as many common parts as possible (although in the end no major part of the gasoline engine could be used in the Toro-Flow). This project was given high priority and several million dollars was allocated to it.⁹⁵ Once more the primary obstacle was achieving good combustion. Two of the engineers who worked on analyzing the purchased Vim engine played significant roles in the Toro-Flow development. After numerous experiments were conducted totalling thousands of man-hours, a commercially satisfactory combustion system was arrived at. Whereas the patented Cummins engines are characterized by a high injection pressure-low swirl rate design, the accused Toro-Flow engines employ a low injection pressure and high swirl rate. The combustion chamber in the Toro-Flow is thus relatively small and deep as compared to a relatively large and shallow chamber in the patented Cummins engines (see Figure 5).⁹⁶ This alternative design of the basic Hesselman-type chamber was well known in the industry prior to the patented engines and the Toro-Flow. However, neither the patent nor the patented engines themselves taught how to design a high swirl chamber. The cautionary warnings in the patent — that the fuel should not impinge on projecting parts and that careful design of the crown and adjacent area of the cylinder was necessary — were helpful tips, but it cannot be said that General Motors copied the patented engines. Despite the warnings contained in the text of the patent, the Toro-Flow has valves that are not recessed, and its combustion chamber more resembles the prior art Ford 172D than the patented engines or the prior art Vera.

95 The record does not indicate how much money was budgeted for this project at GMC Truck and Coach, Engineering Staff or GM Research, but the proposed project at Detroit Diesel Engine Division was budgeted for $2.8 million.

96 In addition, the Toro-Flow injector has four holes, the injectors of the patented engines have eight; the Toro-Flow has two valves, the patented engines, four.

Apart from the design of the combustion system, the differences in construction between the Toro-Flow and the patented Vim are insignificant.

E. Commercial Success.

The patented engines are purchased from Cummins by all major truck manufacturers except General Motors. These engines are commercially attractive to truckers because of their reduced weight as compared to conventional in-line engines.⁹⁷ Each pound of engine weight saved offers the truck operator a potential revenue gain of about one dollar a year from the additional payload that can be carried, the maximum payload being largely determined by state and federal statutes.⁹⁸ However, there were other important factors at work that influenced truckers in deciding whether or not to purchase the patented engines. One was their horsepower requirements.⁹⁹ Another advantage was the dual use of some of the cylinders as compressors for unloading; this was first offered in the patented engines in 1964 and later made available in the Cummins NH-250 in-line model.¹⁰⁰

97 Tr. 760-62 (Shipley); Tr. 807 (Freele).

98 Tr. 760-62 (Shipley); Tr. 805 (Freele).

99 Mr. Freele, of the Schewerman Trucking Co., testified that he had once used the V6-200 engine in twelve or thirteen of his tractors but this had proven unacceptable, so he replaced them with V8-235 units.

100 Tr. 772 (Shipley); Tr. 808-11 (Freele). Mr. Freele's company now has on order NH-250 in-line units in order to take advantage of their potentially greater capacity for unloading.

The sales figures to date are inconclusive as an indicator of commercial success. Between 1961 and 1965, the patented engines accounted, on the average, for only six percent of the total number of open-chamber diesel engines sold by Cummins. The Val/Vale models first hit the market in volume in 1966, but in the 1966-67 period, only about five percent of Cummins' sales were of the patented engines.¹⁰¹

101 Tr. 745-46 (Boll).

The response of experts in the field to both the patented engines and the Toro-Flow has been rather muted. Aside from the expectable puffing with which Cummins introduced the Vim/Vine and General Motors the Toro-Flow, this record contains only reports that one or two experts said they were particularly impressed.¹⁰²

102 The language used by both Cummins and General Motors in advertising their respective new short-stroke engines as innovations is practically the same as that used by Ford in advertising its 172D engine, and by Cummins in introducing its prior art Vera engine. As this Court has stated supra (n. 56) little or no weight is attached to advertising claims of this nature.

VALIDITY

1. Obviousness.

The threshold question before this Court in determining whether the defendants have met their burden of establishing the invalidity of the patent is: What did Reiners invent? He most certainly did not invent the short-stroke internal combustion engine. The historical trend of gasoline engine technology was towards short strokes, and gasoline engines with stroke-to-bore ratios well below 0.9 were in widespread commercial use before 1960. And those gasoline engines employed connecting-rod-to-bore and deck-height-to-bore ratios below the values of 1.75 and 2.8. The patent claims, of course, are directed to a certain type of internal combustion engine — the open-chamber diesel engine. But in arguing that the prior art gasoline engine technology does not make obvious the engines described by the claims, the plaintiff can point only to the differences in the combustion systems, for their kinematic operations are fundamentally identical. Therein lies plaintiff's dilemma: On the one hand, Cummins cannot successfully argue that the solution Reiners used to solve the combustion problem makes a great departure from the prior art diesel technology, since it is the same as was employed by him in the prior art Vera engine,¹⁰³ and is also found in the confines of the other prior art (e.g., in the Briling and Judge articles). On the other hand, if Cummins argues that Reiners' solution does not mark a great departure from the prior art, then there is no patentable invention under 35 U.S.C. § 103, because there is only an engine with parameters well known in the prior gasoline art and a combustion system well known in the prior diesel art. This simple combination of elements is obvious. Therefore, the result of the combination is not patentable. Great A P Tea Co. v. Supermarket Equipment Co., 340 U.S. 147, 71 S.Ct. 127, 95 L.Ed. 162 (1950); Servo Corp. of America v. General Electric Co., 337 F.2d 716 (4th Cir. 1964). See Walker on Patents § 110 (Deller Ed. 1964). Cf. Marston v. J.C. Penney Co., Inc., 353 F.2d 976 (4th Cir. 1965).

103 If plaintiff adopted this position, its suit would also fail for the reason that the text and drawings were insufficient to teach a designer skilled in the art how to build an operative engine.

The plaintiff seeks to avoid the two horns of this dilemma (i.e., the prior art, on the one hand, and the lack of any new departure on the other hand) by emphasizing the uniqueness of the three parameters of the claims working together. Of course, they worked equally well, and in conjunction, in the prior art gasoline engines, yielding exactly the same benefits. In the prior art diesel engine technology itself, the difference between the parameters of the claims and the parameters of either the prior art Vera or the prior art Ford 172D engine is insignificant. In addition, there were a number of prior art diesel engines on the market with stroke-to-bore ratios between 0.9 and 1.0.

It is the claims — not the drawings or the patented engine — that define the scope of the patent, Wayne Knitting Mills v. Russell Hosiery, Inc., 400 F.2d 964 (1968), and the claims are measured against the prior art. This Court has found that:

(1) In any engine the connecting rod length and deck height are made as short as possible;

(2) A reduction in the stroke of an engine enables the designer, by using standard practice, to reduce both the length of the connecting rod and the deck height;

(3) The difference in operation between an engine with parameters slightly outside the claims and one with parameters slightly within is insignificant;

(4) For a stroke-to-bore ratio just within the claims, by using a standard value for the L/R ratio, a diesel engine designer would arrive at a connecting-rod-to-bore ratio within the claims; and a deck-height-to-bore ratio within the claims would follow by standard design; and

(5) The prior art contained detailed analyses and designs, in both the Briling articles and the Vera and Ford 172D engines, of a workable combustion system for an engine described by the claims.

Given these findings of fact, this Court holds as a matter of law that the subject matter of the patent was obvious against the background of either the Vera or the Ford 172d 35 U.S.C. § 103; Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966); Heyl Patterson, Inc. v. McDowell Co., 317 F.2d 719 (4th Cir. 1963); Triumph Hosiery Mills, Inc. v. Alamance Industries, Inc., 299 F.2d 793 (4th Cir. 1962). An observation by the Supreme Court in 1875 — when technology was yet in an embryonic stage — is quite appropriate to this case:

* * * [A] mere carrying forward of new or more extended application of the original thought, a change only in form, proportions, or degree, the substitution of equivalents, doing substantially the same thing in the same way by substantially the same means with better results, is not such invention as will sustain a patent. [Smith v. Nichols, 88 U.S. 112, 119, 22 L.Ed. 566 (1875).]

There is, of course, a presumption of validity set forth in 35 U.S.C. § 282. But where, as here, the Patent Examiner did not cite the most relevant prior art, that presumption is at the very least weakened. Blumcraft of Pittsburg v. Citizens of Southern National Bank, 407 F.2d 557 (4th Cir. 1969); See Heyl Patterson, Inc. v. McDowell Co., 317 F.2d 719, 722 (4th Cir. 1963). Here, as in Blumcraft, "[t]he prima-facie presumption of a patent's validity [ 35 U.S.C. § 282] can be given little weight." Further, the obviousness of the patent is clearly established by the evidence and would in any case overcome that presumption.

The plaintiff has focused much of his argument with respect to defendants' section 103 defense on the "long-felt need" of the industry for the invention, the industry's response to it, General Motors' own unsuccessful history in the field, commercial success, etc. These factors are what the Supreme Court has called "subtests" of patentability. It is true that they are relevant to the issue of patentability and, in close cases, can be of persuasive effect. United States v. Adams, 383 U.S. 39, 51, 86 S.Ct. 708, 15 L.Ed.2d 572 (1966) (operating advantages); Otto v. Koppers Co., 246 F.2d 789, 799 (4th Cir. 1957) (commercial success); Colgate-Palmolive Co. v. Carter Products, 230 F.2d 855, 860-862 (4th Cir. 1956) (lack of success by others over a lengthy period, plus hard to come by). But where, as here, the facts establish convincingly that the invention was obvious against the background of the relevant prior art, secondary tests cannot be controlling. Graham v. John Deere Co., supra; Great A P Tea Co. v. Supermarket Equipment Co., supra; De Forest Radio Co. v. General Electric Co., 283 U.S. 664, 51 S.Ct. 563, 75 L.Ed. 1339 (1931). In any event, the facts in this case, when applied to the secondary tests of patentability, do not establish an innovative quality to the Reiners patent. The commercial success of the patented engines has not been at any point in time outstanding. Although General Motors had unsuccessfully attempted to construct a commercially acceptable short-stroke diesel engine, in none of their projects was there a comparable effort in priority or man-hours of either the patented engines or the Toro-Flow. Also, the prior art diesel technical literature contained several articles which weighed the advantages and disadvantages of going oversquare, but none of those articles expressed any long-felt need. Nor did many experts in the field give widespread acclaim to Reiners' achievements; and, while one or two experts said that they were impressed by the patented engines being able to operate so well at a stroke-to-bore ratio of 0.75 — which, it should be noted again, was well below the upper ratio of the claims of the patent — there was a notorious silence on the significance of the three ratios acting in combination. Compare Power Curbers, Inc. v. E.D. Etnyre Co., 298 F.2d 484, 493 (4th Cir. 1962).

2. Other Statutory Defenses.

In order for a patent to be valid, the invention must satisfy each of the separate statutory requirements of utility, novelty and non-obviousness under 35 U.S.C. § 101-03; and the patent specification and claims, pursuant to 35 U.S.C. § 112, must clearly define and point out, show how to construct, and state the best mode for reducing to practice, the invention. The failure of the invention to pass the tests under any one of those sections is fatal to its patentability. United States v. Adams, supra at 48, 86 S.Ct. 708. The conclusion that the engines described by the claims, when read against the background of the prior art, were obvious to one skilled in the art therefore conclusively establishes invalidity. It seems appropriate, however, in order to round out discussion of this case, to comment on the other statutory tests.

Insofar as the section 101 test is concerned, the utility of the engines made under the Reiners patent has not been seriously disputed by the defendants and is clearly demonstrated by the manufacture, use and sale of the patented engines, as well as of the accused engines. Of course, courts do not try to quantify degrees or measures of utility; it is enough to say that the patented engines have shown themselves to be at least partially capable of accomplishing their claimed functions. 1 Walker on Patents, supra at 492-504.

With regard to the section 112 test of definition, this Court has no difficulty in concluding that the patent claims clearly define and point out the asserted invention. The use of parameters with "non-critical" upper limits, but with no floor, is common to many patents in the internal combustion field issued by the Patent Office ( supra at 77). Further, the open-endedness of the form has not in and of itself seemingly been held fatal by the courts in considering patents in other areas. See, e.g., Lever Bros. Co. v. Proctor Gamble Mfg. Co., 139 F.2d 633 (4th Cir. 1943); E.I. DuPont DeNemours Co. v. Glidden Co., 67 F.2d 392 (2d Cir. 1933). Nor is the presence of a "critical" limit required. See, e.g., Mineral Separation v. Hyde, 242 U.S. 261, 37 S.Ct. 82, 61 L.Ed. 286 (1917); Sun Ray Gas Corp. v. Bellows-Claude Neon Co., 49 F.2d 886 (6th Cir. 1931).

The specification, accompanied by the drawings, essentially told prospective designers that the basic prior art Hesselman system adequately solved the combustion problem of the short-stroke diesel. This satisfies the section 112 requirement that "the best mode contemplated by the inventor for carrying out his invention" be set forth. A more troublesome question is whether enough information was revealed to meet the section 112 requirement of teaching one skilled in the art how to reduce the patent to practice. It is, of course, not necessary for the patentee to have predicted and described every possible variation. See McCullough Tool Co. v. Well Survey, Inc., 343 F.2d 381, 395 (10th Cir. 1965), cert. denied, 383 U.S. 933, 86 S.Ct. 1061, 15 L.Ed.2d 851 (1966). But where, as here, the patentee has singled out a particular problem as being the sole reason for the unsatisfactory results of previous commercial activity in a particular field, he should be held to the duty of providing a specific, informative solution to that problem. The Reiners patent, while highlighting the necessity of overcoming the combustion problem in short-stroke design, seemingly goes no further than to illustrate the well known Hesselman chamber. The prospective designer of a short-stroke engine is still obliged to use his own resources and skill and to engage in very extensive experimentation in order to solve the combustion problem. The only method by which any working parameters can be derived from the patent is to scale the accompanying drawings, but patent drawings are not meant for that purpose, although they are often helpful in illustrating the principles of the patent. Otto v. Koppers Co., 246 F.2d 789, 797 (4th Cir. 1957). And even if it were sufficient to leave it up to others to scale the drawings, the prospective designer is only given a starting point. He would still know nothing about injector location, pressure, etc.; nor would he have knowledge of swirl and swish effects. Indeed, it is pertinent to recall that General Motors invested a huge amount of time to solving the Toro-Flow's combustion problem. On balance, this Court does not believe that the patent meets this requirement of section 112 — that sufficient information be revealed to enable one skilled in the art to reduce the patent to practice. Therefore, the patent is invalid for that reason also.

There remains for consideration the statutory test under section 102 requiring novelty. At issue is whether the Briling articles anticipate the patent. Section 102 provides:

A person shall be entitled to a patent unless —

* * * * * *

(b) the invention was patented or described in a printed publication in this or a foreign country or in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States * * *. [Emphasis supplied.]

This statute on its face draws no distinction between a prior article published in the United States or in a foreign country. Cummins urges, however, that the degree of specificity required for anticipation should be greater for a foreign publication, and there are cases which have so held. See Baldwin-Southwark Corp. v. Coe, 76 U.S.App.D.C. 412, 133 F.2d 359, 365-66 (1942), and cases cited therein. On the other hand, in other cases there is no discussion of any such distinction. Marconi Wireless Telegraph Co. of America v. United States, 320 U.S. 1, 55-57, 63 S.Ct. 1393, 87 L.Ed. 1731 (1943); Dymo Industries v. Com-Tech, 391 F.2d 335 (9th Cir. 1968). When one is faced with a single article from an obscure foreign journal, which may be for practical purposes inaccessible in the United States, it would not appear to serve any purpose of the patent laws to accord it the same weight as an article which is reasonably available to the industry. But that reasoning hardly extends to six articles of a reputable scientist which were published in Soviet technical journals, all within six years of the date of the patent application, and all available in numerous American and British technical libraries. Because the Briling articles were reasonably accessible to the industry, this Court believes that it is proper to apply the same test of anticipation as in connection with any ordinary American publication. That test is that in order to defeat a patent, the prior publication must exhibit a substantial representation of the invention in such full, clear and exact terms that one skilled in the art may make, construct and practice the invention without having to depend on either the patent or on his own inventive skills. Seymour v. Osborne, 78 U.S. 516, 555, 20 L.Ed. 33 (1870), which itself involved a foreign publication; Rich Products Corp. v. Mitchell Foods, Inc., 357 F.2d 176, 180 (2d Cir. 1966), cert. denied, 385 U.S. 821, 87 S.Ct. 46, 17 L.Ed.2d 58 (1966); Bros Inc. v. W.E. Grace Mfg. Co., 351 F.2d 208, 212-213 (5th Cir. 1965), cert. denied, 383 U.S. 936, 86 S.Ct. 1065, 15 L.Ed.2d 852 (1966); Ballantyne Instruments Electronics, Inc. v. Wagner, 345 F.2d 671, 673-674 (6th Cir. 1965). This Court's detailed findings of fact, supra at 83-85, compel the legal conclusion that this test is met by the Briling articles. By employing standard design practice, a person skilled in the art would have, from the information given by Briling, constructed an engine reading exactly on the claims of the Reiners patent. Considerably more information, particularly with regard to the design of the combustion system, is disclosed in those articles than in the Reiners patent. The Reiners patent therefore fails for want of novelty as well as for the other reasons stated in this opinion.

INFRINGEMENT

The finding of invalidity of the patents in suit controls the legal determination of this case. It has been suggested, however, that it is preferable procedure to assume validity and decide the question of infringement. Mabs, Inc. v. Piedmont Shirt Co., 368 F.2d 570 (4th Cir. 1966); Triumph Hosiery Mills, Inc. v. Alamance Industries, Inc., supra at 795. The claims of the patent clearly read on the accused engines. Although there are some differences between the construction of the accused and patented engines, those differences are insignificant. It is not necessary for the plaintiff to show an intent to infringe in order to establish infringement. Thurber Corp. v. Fairchild Motor Corp., 269 F.2d 841, 845 (5th Cir. 1959); see Allen v. Standard Crankshaft Hydraulic Co., 323 F.2d 29 (4th Cir. 1963). This Court therefore concludes that, if the patent were valid, it would be infringed by the defendants. However, for the reasons set forth in this opinion, this Court finds as a fact and concludes as a matter of law that the patent is invalid. Therefore, judgment is being entered for the defendants. It is so ordered.