Opinion
No. 61 Civ. 3644.
April 4, 1967.
Kane, Dalsimer, Kane, Sullivan Smith, New York City, for plaintiff, David S. Kane and Haynes N. Johnson, Stamford, Conn., Robert C. Brown, Jr., and John M. Mann, Chicago, Ill., of counsel.
Davis, Hoxie, Faithfull Hapgood, New York City, for defendant, John Hoxie, Cyrus S. Hapgood and George E. Faithfull, New York City, of counsel.
OPINION
In challenging the validity of a patent, advocates sometimes assert facetiously that the patentee, instead of patenting an invention, invented a patent. The occasional aptness of this cliché is illustrated by the present litigation.
The railroad freight car cushioning apparatus that is the subject of the patent in suit differs materially from its predecessors only in the respect that the cushion specified has a longer stroke or yield. Having found that this sole material difference is a distinction without inventive significance, the Court has concluded — for the reasons more particularly set forth in the course of the opinion — that the patent is invalid.
This action, commenced by the filing of a complaint on October 16, 1961, is for the alleged infringement of United States Letters Patent No. 3,003,436 on a "Method Apparatus for Protecting Vehicle Loads" issued in the name of William H. Peterson on October 10, 1961. Plaintiff, a Delaware corporation with a place of business in New York, is and has been, since the issue date thereof, the owner of the Peterson patent. Defendant is a New Jersey corporation with its principal place of business in New York. The Court has both in personam and subject-matter jurisdiction.
The Peterson patent was granted on an application of December 3, 1959. That application was a continuation-in-part of an earlier application of March 5, 1959, the disclosure of which, to the extent that it is not inconsistent, is expressly incorporated by reference in the subsequent application and in the patent.
The patent itself is directed to the protection of lading, while it is being transported in or on a railway freight car, against damage arising from longitudinal impacts to the car such as those which commonly occur during switching operations in classification yards. Moreover, the patent states that the invention has broader aspects in that "its benefits can be applied to various types of passenger-carrying vehicles as well as freight-carrying vehicles."
Reduced to simple terms, the subject of the patent is a cushion or shock absorber of stated characteristics which is interposed at some point between the car body and the car couplers. The cushion serves to give some degree of protection to the vehicle and its contents from impacts received at the couplers. Since cushions of various kinds had been used for many years in railway cars to serve the same basic purpose, the fundamental questions arising in this litigation revolve around the stated characteristics of the cushion described in the patent claims. As claimed, the cushion is of the energy-dissipative type, and is identified by its travel stroke, closure time and force-travel characteristic.
The defendant, one of the plaintiff's competitors in the field of railroad car building, is charged with infringement of the Peterson patent by reason of its admitted manufacture, use and sale of freight cars of the known "cushion-underframe" type employing one or another of three styles of hydraulic cushion. Those styles, each of a different make, are:
a) The "Freight Saver," formerly made by the Bendix Products Division of Bendix Corporation but now made by defendant itself since its acquisition, on or about November 15, 1962, of the freight-car cushion business of Bendix. The Freight Saver is made in two sizes with respect to the nominal length of cushion travel: 20 inches and 30 inches.
b) The "Hydra-Cushion," sold by Hydra-Cushion, Inc., which has a nominal length of cushion travel of 20 inches.
c) The "Shock-Control" unit, sold by Keystone Railway Equipment Co., which has a nominal length of cushion travel of 20 inches.
Plaintiff relies upon 13 of the 27 claims of the Peterson patent. It divides the selected claims into three groups designated A, B and C, as follows:
A claims: Nos. 1, 8, 9, 12, 15, 16, 18, 21 and 27.
B claims: Nos. 7, 17 and 26.
C Claims: No. 20.
Plaintiff takes claim 1, and defendant takes claims 1 and 21, to be typical of the A claims. Plaintiff takes claim 7, and the defendant takes claims 7 and 26, to be typical of the B claims. Claims 9 and 27 of the A group and claim 26 of the B group — which are limited by their terms to a cushion having a "stroke" or travel length of about 30 inches, or to a cushion with characteristics equivalent to a 100% efficient cushion having that length of stroke — are relied upon only in support of the charge against the 30-inch Freight-Saver. The C claim, No. 20, is relied upon only against the Hydra-Cushion.
In addition to denying the charge of infringement, defendant raises several issues of patent validity. Specifically, defendant contends that:
a) The patented subject is an old combination, has no material novelty and is anticipated in substance ( 35 U.S.C. § 101, 102);
b) The patented subject was not inventive but was "obvious," within 35 U.S.C. § 103; and
c) The patent fails in material respects to comply with 35 U.S.C. § 112, both as to its disclosure and its claims, and especially as it is asserted in this action.
I. BACKGROUND TECHNOLOGY
A familiarity with some basic technology of railway freight cars and some elementals of railroad operations is helpful to an understanding of the issues here in controversy.
In simplified form, a conventional or standard freight car consists of: a car underframe mounted on the wheeled trucks; a car body mounted, in turn, on the underframe; couplers projecting from each end of the underframe; and draft gear units, which are friction or rubber shock absorbers of less than three-inch yield, interposed between each coupler and the underframe. Thus, a push or pull received by the coupler is applied endwise to the underframe and car body through the draft gear.
Freight cars receive longitudinal impacts in being coupled together. During train operation, impact forces are also applied at the couplers when the slack present in the couplers and draft gears permits the coupled cars to be pulled somewhat apart or squeezed somewhat together. A severe, though common, impact occurs in the assembly of cars into trains at classification yards when a car is set in motion to run down a track into the end car of a previously assembled string of cars, which offers little "give."
To protect railway cars and their contents, railroads sought for many years to limit car speeds to 4 miles per hour in the coupling of cars. It was and is generally accepted that, when the car speed is held below this rate, the impact at coupling is sufficiently cushioned by the conventional draft gear so that it does not cause damage to the cars or to most of the many kinds of lading, of diverse degrees of fragility, that are carried. Above this speed, however, the shock absorbing capacity of the draft gear is generally insufficient; and there may be damage to some forms of lading resulting from the higher force applied to the car following closure of the draft gear. Because the energy of the striking car, and therefore the severity of the impact, increases as the square of the speed, the probability of damage is multiplied as the speed is greater.
Despite attempts by railroads to keep coupling speed down, increased traffic demands and other factors resulted in frequent higher coupling speeds including speeds up to and above 10 miles per hour. It was well known that, because of their modest cushioning capacity, conventional draft gears could not generally provide adequate lading protection at speeds higher than 4 miles per hour. Although the cushioning capacity could be increased by increasing the travel of the draft gears, such a solution was proscribed in practice because it would increase the extent of slack in trains, thereby creating difficulties in train operation.
In recognition of this need for greater cushioning capacity, the "cushion-underframe" car was introduced in the late 1920's. In this car, the couplers are mounted, frequently through draft gears, on the ends of a sliding sill, a long and heavy member which slides lengthwise in a channel in the center sill which is the principal longitudinal beam of the underframe. A cushion forms the connection between the sliding sill and the center sill of the underframe, so that an impact received at the coupler at either end of the sliding sill is applied through the cushion to the underframe and the attached body. Thus, supplemental cushion yield can be provided without introducing additional slack between the cars of a train (sometimes denominated the "bellows" effect) because the through connection of car to car is by way of the rigid sliding sills commonly carrying only conventional draft gears and couplers.
To understand the manner by which the cushion affords protection to the lading, it is necessary to consider the sequence of physical events which occur during a coupling impact. To simplify the explanation, the case to be described will involve a cushion-underframe car (containing the lading under consideration) which is struck by a moving uncushioned car while the cushioned car is at rest but free to move. While this is not the most common impact situation, it is a satisfactory model for understanding the sequence of events. When the coupler of the moving car makes contact with the standing cushion-underframe car, the impact is received by the coupler at one end of the sliding sill. The immediate result is that the coupler starts to move inward. If there is a draft gear interposed between the coupler and the sliding sill, the initial motion of the coupler is resisted yieldingly by the friction elements or the compressible rubber elements of the draft gear. The resistive force created in the draft gear acts in both directions, tending to set in motion the sliding sill while also starting to retard the striking car. The inertia of the sliding sill and the frictional resistance between the sliding sill and the walls of its channel in the center sill offer additional resistance to motion of the sliding sill.
This illustrative model is similar to the setup used by Peterson in his experimental impact tests in that a single car is run into a single standing car which is free to roll. In Peterson's experiment, however, the cushion was mounted on the end of the impacting car. See p. 289, infra.
As soon as the sliding sill is set in motion, it pushes on one end-element of the underframe cushion located between the sliding sill and the car's underframe, and causes that cushion element to move toward the other end, against the yielding resistance offered by the interposed cushioning medium. During the resulting travel of that moving element of the cushion, the resistive force developed by the cushion acts not only to resist the motion of the sliding sill (and thereby to further retard the striking car which is pushing on the sliding sill) but also to cause the other end-element of the cushion to push on the underframe of the struck car and set it and the attached car body in motion. A portion of the energy of the striking car is thereby expended in overcoming the resistance of the cushion and in setting the struck car in motion.
During this period of acceleration of the struck car by application of the cushion force to the underframe, lading in that car is also brought up to the same speed by reason of whatever connection it has to the car body. This occurs with or without relative motion between the lading and the car body depending upon the connection between the two and the abruptness or gradualness of the speedup of the car body.
The phenomenon by which lading is damaged as a result of coupling impacts is complicated. As a fundamental proposition, it can be stated that a piece of lading becomes damaged when a force is applied to it which exceeds its resistive strength and either defaces, deforms, or fractures the lading. The difficult question is how such a force comes to be applied. The lading carried by a general purpose freight car varies considerably in rigidity, fragility, packaging, weight and volume of individual pieces, weight and volume of total load, and in arrangement and bracing within the car body. As the car body is accelerated by the cushion force on the underframe, the relative motion or tendency to relative motion between the interior surfaces of the car body and the lading results in the application of forces on the lading.
While potentially damaging forces can be developed between pieces of lading or between one or more pieces of lading and the floor or a side wall of the car body, the primary source of damage is the forces applied to the lading by the endwall of the car body at the struck end of the car. In the illustrative impact previously described, if the cushion force is comparatively high, it both retards the striking car and accelerates the struck car more quickly. Conversely, if that cushion force is comparatively low, the change of speed of each car is accomplished more gradually.
The forces applied to the lading at the endwall of the car body at the struck end of the car (termed "lading forces" in the patent) are greater or less depending on how abruptly or how gradually the car is brought up to its terminal speed. The particular relationships between lading forces, car body acceleration, and cushion force were matters given much consideration in the present litigation.
The lading forces are affected by various characteristics of the lading and the arrangement of lading within the car body. Moreover, to the extent there are frictional connections between the car body and the lading, frictional forces are introduced which tend to make the lading move with the car. These frictional forces reduce the relative motion or tendency to relative motion between the lading and the car body, thereby reducing the endwall forces which otherwise would result in the absence of friction. The magnitude of the frictional forces, in turn, is affected by such factors as the arrangement of the lading within the car body, the characteristics of the lading and the car body surfaces, and the nature and extent of the contact between those surfaces. Furthermore, if the product of the cushion force and the distance between its point of application and the center of gravity of the loaded car is sufficiently great, there will be a tendency for the struck end of the car to drop while the opposite end rises. This vertical bounce will disturb the lading, thus affecting the frictional contact and hence the magnitude of the frictional forces.
From this brief explanation it is manifest that the exact relationship between car body acceleration and endwall force on the lading is impossible of calculation as a theoretical matter. Nevertheless, it would be virtually beyond question, as an engineering matter, that the endwall force on the lading is greater as the acceleration of the car body is greater; and that the car body acceleration is greater, in turn, as the cushion force producing the change in speed is greater. The elemental conclusion from the foregoing is that the desired action of the cushion is to apply a force low enough to keep the car body acceleration low enough to hold endwall forces on the lading below the level at which the particular kind of lading under consideration will become damaged.
Consistent with these propositions, it has been and still is a common practice to measure the coupler force (the force developed at the meeting of the couplers which is approximately the same as the cushion force except while the inertia of the sliding sill is being overcome) and also the car body acceleration — both values being an index of potential damage to the lading during the coupling impact.
Apart from the criterion of cushion force during operation of the cushion, adequate cushioning against a longitudinal impact requires that the extent or distance of the cushion travel be long enough so that the cushion does not close and "go solid" before the impact energy has been spent through absorption in the cushion and acceleration of the struck car. If the cushion closes before this has been accomplished, the unspent energy is then transmitted to the underframe and car body through an uncushioned connection which develops a higher force productive of a higher acceleration and higher resulting lading forces.
Having explained the function of a cushion in relation to the forces acting upon the lading, the Court now turns to an examination of actual cushion structures. Described simply, a cushioning device is a double-ended structure consisting of three basic elements: (1) a movable solid element at one end which receives the impact; (2) a relatively fixed solid element at the other end; and (3) a cushioning medium between these two ends which offers a yielding resistance to movement of the movable element toward the other end-element when the movable element receives an impact.
A cushion is commonly characterized by its force and its travel. In engineering terms, the "work" it does — or the energy it absorbs — when impacted is the product obtained by multiplying its average force by the distance over which it travels. The force-travel characteristic of a cushion is conventionally portrayed by a plot in which the vertical scale is in units of force and the horizontal scale is in units of travel distance, giving a "curve" representing the value of the force from point to point during the travel. The area under this curve represents the "work" done or the energy absorbed. To absorb a given amount of energy, a cushion may exert a high force over a short travel or a lower force over a correspondingly longer travel — the area under each force-travel curve then being the same.
A cushion which exerts a uniform force throughout the length of its travel may be termed a constant-force cushion. Such a cushion is said to have absorbed the given energy with "minimum travel," as compared to a cushion which exerts a lower force at some points in the course of its travel and, therefore, requires a longer distance of travel to absorb the same energy.
The "efficiency" of a cushion is defined and calculated as the ratio of (a) the area under its force-travel curve, representing its "work," to (b) the area of a rectangle defined by the cushion's maximum force and travel length, representing the work that it hypothetically could do in the same travel distance if its force throughout the course of its travel were the maximum it actually exerted only over a part of its travel. It follows inexorably from this definition of "efficiency" that only a constant-force cushion will exhibit an efficiency of 100 per cent, whereas any cushion which exerts a varying force during its travel must have a lower efficiency.
While all functional cushions absorb a substantial portion of the impact energy, they operate differently in disposing of this energy. In terms of energy disposition, cushions may be divided into two broad categories, with the knowledge that most practical cushions are likely to display characteristics of both groups. The energy-dissipative cushion dissipates most of the absorbed energy, generally in the form of heat; whereas the energy-conservative cushion stores most of the absorbed energy as potential energy within the cushion, releasing that energy to the dynamic system as the cushion returns itself to its initial position.
A number of different cushioning media have been used in the cushions of cushion-underframe cars. Compressible coil springs had been used in the first such cars. Early in the 1950's, plaintiff began producing cushion-underframe cars employing compressible rubber as the cushioning media. In the mid-1950's, the 10-inch travel "Hydra-Cushion," an energy-dissipative cushion, was introduced. This cushion is composed of solid metal plates slidable one on another but held together in a way to develop a frictional resistance to relative motion, with a hydraulic unit to control the resistive force.
The claims of the Peterson patent, which will be discussed more fully in the course of this opinion, call for an energy-dissipative cushion of certain stated characteristics, including a particular force-travel characteristic. All of the defendant's allegedly infringing railway cars employ cushions of the energy-dissipative type, and, more particularly, utilize hydraulic elements in achieving the cushioning action. Because important questions of patent validity and infringement turn on the claimed force-travel characteristic, a familiarity with the characteristics of hydraulic elements is necessary.
Hydraulic cushions operate on the principle that a resistive force is created — and energy is dissipated as heat — when a liquid is forced through a restricted orifice. Commonly, this principle is put to use by means of a body of liquid contained within a cylinder in which there is a movable piston. A restricted orifice of one form or another is provided through which the contained liquid is forced from one side of the piston to the other when relative motion occurs. The restriction of that flow through the orifice creates a force which yieldingly resists the relative motion of piston and cylinder, dissipating as heat a part or all of the energy of the impact that causes such motion.
The resistive cushion force created by such a restrictive orifice is greater as the velocity of impact — and resulting velocity of relative motion of piston and cylinder — is greater. More specifically, because the resistive force increases in proportion to the square of the speed — as does the impact energy — a cushion embodying this hydraulic resistance principle accommodates itself to the severity of the impact by a given mass, attaining approximately the same travel over a broad range of speeds. The initial orifice area is chosen so as to give a desired range of resistive forces over the range of impact speeds anticipated.
If the orifice area of the hydraulic cushion is unchanged during the cushion travel, the resistive force decreases as the velocity of travel decreases due to the cushion's resistance. It is conventional, therefore, to vary the orifice area — as the travel proceeds — by some device such as a metering pin of tapered contour which projects through the orifice and partakes of the relative motion of the piston and cylinder to gradually close off the orifice. Because the resistive force has an inverse relationship to the orifice area, the gradual reduction of the orifice area tends to compensate in whole or in part for the decreasing velocity.
The force-travel characteristic of a cushion utilizing hydraulic resistance is dependent not only upon the impact velocity but on the masses involved in the impact. While metering commonly is used to compensate for the change in velocity during travel, no similar mechanism has been devised to adjust the force level for the variation in the impacting masses encountered in ordinary operation. Thus, the initial orifice size and the metering device must be designed on the basis of some particular assumed set or range of masses, with due regard for the variation in force-travel characteristic when masses other than those anticipated are encountered. Stated generally, the variation will be in the direction of an increasing resistive force over the latter part of the travel when the actual effective mass is greater than that assumed, while a tendency to decrease will be present when the actual mass is smaller.
Three styles of cushion employing hydraulic resistance are involved in this litigation: the straight hydraulic cushion, the hydropneumatic cushion, and the hydraulically-controlled friction cushion. In the straight hydraulic cushion, such as the accused Keystone cushion and plaintiff's production cushion, the cushion force is developed entirely by hydraulic resistance at the orifice, except for the small resistance of the return spring which is used to restore the moving part to its normal position. In the hydropneumatic cushion, represented by the accused ACF (Bendix) cushions, a supplemental pneumatic unit — which compresses air during the cushion travel — is used to return the cushion to its initial position. Finally, there is the hydraulically-controlled friction cushion, such as the accused Hydra-Cushion, whose operation already has been described. In this cushion, a major part of the cushion force is developed by friction between the interleaving sliding brake plates. The frictional resistance force is a function of the pressure between these plates which is controlled by a hydraulic unit arranged so that its piston is displaced as the travel of the friction plates proceeds. In this fashion, the cushion force is regulated by the hydraulic unit, which also serves to dissipate a portion of the impact energy.
See p. 285, supra.
One final point of background information worth noting is suggested by the lengthy explanation which has preceded it. A freight car cushion must be designed in reference to two fundamental matters: (1) the level of lading protection desired (maximum lading force), which may be said to fix the tolerable cushion force; and (2) the range of coupling speeds and impacting masses over which that level of protection is to be afforded, which fixes the maximum impact energy anticipated, and, therefore, in relation to the cushion force limitation, fixes the cushion travel for a desired force-travel characteristic.
Because ladings differ both in characteristics and in arrangement within the car, and because higher coupling speeds and mass situations are encountered less frequently, no absolute judgment as to design criteria is possible. Consistent with common engineering practice, a compromise must be achieved by considering probabilities and costs. It is, therefore, inherent in the design of a freight car cushion that there is always a question of how much cushioning is enough.
II. PATENT DISCLOSURE AND CLAIMS
While the Peterson patent discusses its subject in terms of the railway car as whole, the parties have stipulated for purposes of this case that the following combination was in public use as described in printed publications prior to June, 1955:
"A car underframe, a draft buffing column (sill) carried by the underframe and capable of longitudinal movement relative thereto, couplers with draft gears operatively connected thereto at opposite ends of the column, a load carrying body supported by the underframe and longitudinally movable relative to the column and couplers, means for supporting said underframe for travel over the rails of a track, and dissipative energy type cushioning means interposed between the underframe and the column for providing cushioning therebetween through a certain maximum distance of cushion travel in each direction." (R. 440)
The patent itself acknowledges the prior use of cushions on freight cars in the cushion-underframe cars with a sliding sill. Two other locations for the cushion in relation to other parts of the car also are suggested. In any event, as it is to be gathered from the disclosure of the patent and the evidence at trial, its purported innovation with respect to apparatus is not in the specific structural features of the cushion or its location within the car; rather, it is in the combination of certain cushion characteristics. Specifically, the patent calls for an energy-dissipative cushion having a cushion travel within a particular range of values, a minimum closure time at a stated impact speed, and a particular force-travel characteristic — namely, 100 per cent — at least in the preferred embodiment of the invention. Although the patent asserts that the prior cushions have not solved the problem of lading damage with high speed impacts, it acknowledges that draft gears and previously used cushions satisfactorily protected lading in low speed impacts.
The limitation to the dissipative type of cushion excludes cushions of the energy-conservative type which store the absorbed impact energy and, therefore, exert a relatively strong recoil after the impact is over. While the patent describes specific examples of two styles of cushion structure — the straight hydraulic cushion and the hydraulically controlled friction unit — it nevertheless states its contemplation that a wide variety of other styles may be used, provided they are of the dissipative type, except for energy stored and used to return the device to its normal position. The patent acknowledges some prior knowledge of hydraulic cushions and their design.
In the Peterson patent, the term "cushion travel" or "stroke" generally means the maximum distance through which the movable element can travel before "going solid" against built-in stops. That travel is not necessarily fully attained under all impacts. As part of its purported novelty, the patent discloses a cushion travel in the range from approximately 20 inches to 40 inches. A "leeway" is stated as being 7 per cent on either side of the 20- to 40-inch range. Whatever may be the basis for this leeway, plaintiff has taken the position in this action that the "leeway" language, as variously stated in the patent, fixes a cushion travel range from 18.6 to 42.8 inches.
The patent also states a correlation between its range of cushion travel and the range of common freight car lengths, concluding that a cushion travel of about 5 per cent of the car length falls, roughly, within the stated range. The disclosure of the Peterson patent expressly recognizes that to the extent that a cushion does not have the preferred characteristic — i.e., substantially constant force-travel and therefore substantially 100 per cent efficiency — the patent's entire range of cushion travel values is to be increased or adjusted correspondingly. Such cushions of lower efficiency but longer travel are considered the equivalents of the preferred cushion of 100 per cent efficiency having a travel in the 20- to 40-inch range.
With respect to cushion closure time, the patent discloses the closure times — calculated by a previously known elementary formula — for a conceptual 100 per cent efficient cushion having adequate cushioning capacity that has a truly constant force throughout its travel and, therefore, gives a uniform deceleration. For such a cushion it discloses that, for an impact at 10 miles per hour, the closure time for a travel of 20 inches is 0.23 second and for a travel of 30 inches the closure time is 0.34 second. In addition, it gives experimentally determined closure times — essentially in accord with the calculated time — for a few impacts with Peterson's straight friction cushion of very close to 100 per cent efficiency.
The Peterson patent does not disclose any cushion closure time for a hydraulic cushion. Moreover, with respect to cushions, of whatever style, having an efficiency less than 100 percent, it discloses neither the closure time, nor how to calculate closure time, nor whether the closure time is greater or less than that of the 100 per cent efficient cushion.
There is no dispute that, for a given impact speed, the closure time of a 100 per cent efficient cushion is directly related to its travel, and is an alternative way of defining the travel in numerical terms. The closure time at a given speed of a cushion of less than 100 per cent efficiency also may be calculated, though the mathematics are considerably more complicated. Nevertheless, the evidence adduced shows that the closure time of such a cushion is invariably longer than that of the 100 per cent efficient cushion when its maximum force is the same and its travel is increased to compensate for its lower efficiency.
In regard to the final significant feature, the cushion's particular force-travel characteristic, the Peterson patent states a preference for a cushion of 100 per cent efficiency — i.e., one having a constant force-travel characteristic. However, it is a fact, recognized in the patent, that a perfectly constant force cannot be attained in practice. Therefore, the preference is qualified in that "a reasonable deviation as a result of practical design limitations can be tolerated." Elsewhere, in a general statement which does not purport to express the patentee's preference, it is said that some variation from a 100 per cent efficient cushion can be tolerated whether it be due to design limitations or "otherwise." This suggests a departure from the preferred force-travel characteristic of 100 per cent efficiency for reasons other than design limitations.
The Peterson patent does not given any numerical range to describe the variation from 100 per cent efficiency that would still be within its preference or optimum force-travel characteristic. The only criteria disclosed are (a) the above references to practical design limitations; (b) a related statement that, as one source of variation, there may be instances in which the cushion may exhibit high force peaks but of such short duration as not to affect the action materially, particularly with respect to the lading forces; and (c) a single plot (curve 305 of Figure 23 of the patent) of the force values prevailing during the working of the straight friction cushion on which are based all of the experimental data of the patent, except data on conventional draft gears. That plot exhibits a rapid attainment of the working cushion force at the onset of the impact, a few early peaks of very short duration and of no effect on the lading force, and a force variation thereafter that is less than 10 per cent above or below the average value.
For its disclosed hydraulic cushion, the Peterson patent gives a formula for the shape of the tapered metering pin in the case where a completely rigid body is being cushioned. The formula is described as a "best starting point" which may be altered "to give a closer approach to the optimum constant force-travel characteristic for a given situation after a few experimental trials." This same description of the hydraulic cushion metering pin states that there is no limit to the possibilities of how the pin might be shaped to suit special situations, thereby suggesting the possibility of a considerable departure from the optimum constant force-travel characteristic for the "given car weight."
The Peterson patent uses the term "constant force type" only in reference to the straight friction cushion used in Peterson's experimental work, although it recognizes that, for a given impacting mass situation, some undefined close approach to a constant force-travel characteristic can be attained with a hydraulic cushion. Prior art publications in the art of hydraulic cushioning taught that such a cushion can be designed to have an efficiency as high as 93 per cent for a particular mass array. This would be comparable to Peterson's experimental friction cushion. Indeed, it is to be noted that all of the experimental results and effects set forth in the patent which are said to attend the use of a cushion travel in excess of 18.5 inches are based on tests involving the straight friction cushion which approached 100 per cent efficiency. A cushion operating at substantially less than 100 per cent efficiency, e.g., one of 60 or 70 per cent efficiency, could not yield the reported results and effects unless the less efficient cushion, in proportion to its lesser efficiency, has a travel which is longer than that particular travel of the 100 per cent efficient cushion. This conclusion follows logically from the relationship between force, travel, and energy in view of the need to maintain the same maximum cushion force while absorbing the same amount of energy.
With specific reference to the claimed results concerning lading protection in an uncompartmented railway car employing the alleged invention, the only experimental findings disclosed in the patent are based on eight impact tests, four at 9.5 to 9.7 miles per hour and four at higher speeds up to 14.2 miles per hour. The eight impact tests were part of a longer series which also tested the effect of compartmenting the car, a known practice for the protection of lading. In each impact test, a freight car, without draft gears and free to move, containing cartoned canned goods was impacted by another car having at its striking end the substantially constant-force, straight friction cushion mentioned earlier. Parenthetically, it may be noted that these impact conditions are neither the most common nor the most severe that are encountered in railroad operations. More frequently in actual practice, a single car or several coupled cars are run into a standing string of previously assembled cars that does not roll away when impacted.
At impact speeds of 9.5 to 9.7 miles per hour, the lading was protected when the friction cushion had travels of 19, 221/16, 25 1/8 and 307/16 inches. The measured force developed between the endwall of the car at its struck end and the adjacent tier of cartons (called the lading force) was in all cases below the value of 2,000 pounds per square foot that was independently found, in a static test of that lading, to be the level at which compressive force damage began to appear.
At higher speeds of 11.6 to 14.2 miles per hour, the same lading was protected when the friction cushion had travels from 26 7/8 inches to 32½ inches respectively. The measured lading force in all four impacts at these speeds likewise was below 2,000 pounds per square foot.
On the basis of certain of these impact tests, the patent presents a curve (302 in Figure 19 of the patent) showing the relation between cushion travel and lading force at 10 miles per hour. It appears from this curve that at 10 miles per hour, under those specific test conditions, the force on that cartoned canned goods lading could be kept below the damage level with a cushion travel of 18 to 19 inches.
An independent static compression test on a species of cartoned bottled goods showed their damage level to be in the neighborhood of 1,000 pounds per square foot. The experimental cushion travel-lading force curve depicted in the patent (curve 302 in Figure 19) indicates that, to hold the lading force below 1,000 pounds per square foot at ten miles per hour, a cushion travel of 21 inches is required. Moreover, other curves depicted in the patent reveal that, under the following conditions, the lading force exceeded 1,000 pounds per square foot: 19 inches of cushion travel at 9.5 miles per hour; 26 7/8 inches of cushion travel at 12.6 miles per hour; 28½ inches of cushion travel at 13.6 miles per hour; and 32½ inches of cushion travel at 14.2 miles per hour.
The patent also details the relation between coupler force and lading force, and the effect on each as cushion travel is increased. Among the points disclosed is the known pattern of reduction of coupler force as greater lengths of cushion travel are employed for a given impact speed. The patent states, without supporting documentation, that it formerly was believed generally that lading force and lading damage are proportional to coupler force. It asserts that, on this generally accepted premise, there was little benefit to be expected from an increase in cushion travel because the reduction in lading force anticipated on the basis of the reduction in coupler force would not be sufficient, for any practical cushion travel length, to hold the lading force below 1,000 pounds per square foot. The disclosed experimental findings, on the other hand, revealed that lading force was reduced by much more than coupler or cushion force as the travel of the straight friction cushion was increased from 19 to about 30 inches. Specifically, the lading force was reduced by 83.4 per cent against a cushion force reduction of 30.6 per cent.
As presented in the patent, this disproportionality would appear to be an effect which is peculiar to cushion travels in excess of about 18.5 inches. However, the patent fails to present any data which would make possible a comparison of forces at shorter travel lengths. In a not entirely clear way, the patent relates this disproportionality to the role of (a) friction within the lading and between the lading and the interior surfaces of the car body, and (b) the inherent stability of tiers of cartoned lading against toppling. The tendered explanation is that, with shorter cushion travel lengths, these effects are of no consequence, whereas with cushion travel lengths above approximately 20 inches they become principal factors in accelerating the lading, thereby reducing the pressures at the endwall which cause damage.
The Peterson patent also deals at length with the relation between cushion closure time and the time required to reach a peak endwall force on compressible lading (denominated maximum compression or compaction time) during an impact. The conclusion stated is that cushion closure time should exceed maximum compaction time to afford protection to the lading. For impacts at 10 miles per hour, this relation is said to obtain for Peterson's experimental cushion at travel lengths of 18.5 inches or more. Some limited test data on maximum compaction times measured during Peterson's impact tests are disclosed in the patent.
The discussion of the subject of compaction time rests on a comparison with impacts involving conventional freight cars employing draft gears as the sole cushioning medium. Above 3 to 4 miles per hour, the draft gear closes before much of the impact energy has been absorbed, causing the impacted car to receive a high uncushioned force and a resulting abrupt acceleration. In that case, the peak force on the lading and the consequent maximum compaction of the lading occur after closure of the draft gear because of the insufficient cushioning capacity.
Another advantage said to accrue when cushion travel exceeds approximately 20 inches is the reduction of vertical bounce during impact. This effect enhances the frictional connection between the lading and the car floor, further reducing the magnitude of the endwall forces on the lading which are the primary cause of damage. Beyond approximately 30 inches of cushion travel the frictional forces are said to have increased to the point where they are doing most of the work in accelerating the lading. The endwall provides 21.4 per cent of the total impulse applied to the lading at 30 inches of cushion travel, compared to 70.6 per cent with a cushion travel of 19 inches.
The "A" claims of the Peterson patent — couched in terms of apparatus — define a combination of conventional freight car components together with a cushioning device of stated characteristics interposed beween the couplers and the car body. As noted earlier, consideration of these claims may be narrowed for purposes of this litigation to the stated characteristics of the so-called energy-dissipative cushions.
Plaintiff takes claim 1 and defendant takes claims 1 and 21 as representative of "A" claims. The pertinent language of claim 1 is a "cushioning device having energy transferring and dissipating characteristics equivalent to a 100% efficient cushioning mechanism having a cushion travel in one direction within the range from about 20 inches to about 40 inches, and in which for an impact of 10 miles per hour * * * the cushioning device will not close in less than about .23 second, said cushioning device having a cushion stroke length for closing in one direction of not more than about 40 inches." The pertinent language of claim 21 is a "cushioning unit * * * having a cushioning closure stroke, for a given impact, equivalent to 100% efficient cushioning device having a cushion closure travel within the range of about 20 inches to about 40 inches * * * said cushioning unit having a closure time for an impact of 10 miles per hour of not less than about .23 second and a cushion stroke length for closing in one direction of not more than about 40 inches."
The form of statement used in each claim refers to two cushions: (1) the cushion which is being defined as the cushion embraced by the claim, the travel range of which is not stated in numerical terms; and (2) the 100 per cent efficient cushion which is included in the statement as the standard of reference, the travel range of which, stated as about 20 to 40 inches (about 30 inches in claims 9 and 27), is the range disclosed in the patent. The principal difference in wording between claims 1 and 21 concerns their respective subject of the equivalence — claim 1 referring to certain energy transferring and dissipating characteristics and claim 21 referring to the travel length or stroke. Nevertheless, because the patent specification discloses only a single basis of equivalence and plaintiff takes claim 1 alone to be typical of the "A" claims, the Court finds that the meaning of these two claims is essentially identical.
In light of the patent's disclosure concerning equivalence and of the meaning given to the "A" claims when they were presented to the Patent Office, and in view of the legal criterion of equivalence which plaintiff invokes here as the criterion for these claims, the necessary and proper interpretation of the statements in the "A" claims concerning equivalence is that the claimed cushion has a travel (stroke) of such length that it dissipates the same energy and transmits energy to the car body at the same maximum force level as does the defined 100 per cent efficient cushion used as a reference standard. Such a cushion would thereby perform the same cushioning function in the same way to give the same result. As the patent recognizes, two cushions differing in efficiency can only be equivalent on that basis if the less efficient cushion has a greater travel.
The remaining uncertainty concerning the meaning of the "A" claims results from the statements that the closure time, at 10 miles per hour, is not less than .23 second (.34 second in claims 9 and 27) and the cushion stroke is not. more than 40 inches. The question is whether these qualifications further define the cushion that is claimed or the cushion of 100 per cent efficiency that is used as a reference standard. While a purely semantic approach would relate the statements to the claimed cushion rather than to the reference cushion, the general imprecision of the claim language and the ambiguous use of synonymous terms within these very claims indicate that the question should not be resolved on a semantic basis.
However, even when considered on the basis of substance, the problem is no less confusing. As noted earlier, the patent discloses that a 100 per cent efficient cushion of 20-inch travel inherently will close in about .23 second (.34 second for a 30 inch travel) at an impact speed of 10 miles per hour if the cushioning capacity is adequate. Indeed, a statement of closure time for a 100 per cent efficient cushion at a given speed is no more than an alternative expression of the length of cushion travel. Therefore, if applied to the "reference cushion," the statement of closure time would be somewhat redundant. Likewise, the stated upper limit for cushion travel of 40 inches would serve no useful purpose if applied to the "reference cushion" since its travel range is already specified as within about 20 to 40 inches. On the other hand, if the stated upper limit is treated as applicable to the "claimed cushion," a question would be posed as to why the patentee limited the travel of his claimed cushion to 40 inches in view of the patent disclosure that "with respect to the upper end of the entire range, the entire range shifts upwardly to the extent that the dissipating energy cushion is not 100% efficient." Nevertheless, the evidence, considered in its entirety, compels a finding that the statements of closure time and upper travel limit are to be related to the claimed cushion rather than to the reference cushion.
The "B" claims, of which claims 7 and 17 are typical, likewise are couched in terms of apparatus and define essentially the same general combination of freight car components as do the "A" claims. The only portions of the "B" claims that require consideration are the statements which define the cushion (a) by the travel range and closure time, and (b) as being "substantially of the 100% efficient * * * type" (claim 7), or as "having force-travel characteristics which approximate a 100% efficient cushion" (claim 17). The parties have agreed that the terms "substantially" and "approximate" are synonymous. In claim 7, a cushion travel range of about 20 to 40 inches is described only indirectly by a statement of minimum closure time of .23 second at 10 miles per hour, and maximum stroke of 40 inches. In claim 17, a range of about 20 inches to about 40 inches is specified, and the corresponding minimum closure time of .23 second at 10 miles per hour is also stated. Similarly, in claim 26, a stroke of about 30 inches is stated, with a corresponding minimum closure time at 10 miles per hour of .34 second.
The "B" claims were described by plaintiff as being narrower than the "A" claims; and the language and substance of the respective claims appear to support its position. The "A" claims do not purport to limit the force-travel characteristic of the "claimed cushion" to within any tolerance of the reference cushion of 100 per cent efficiency. On the other hand, the "B" claims clearly are directed to achieving some reasonable approximation of the preferred constant force-travel characteristic.
The single "C" claim, No. 20, again states essentially the same general combination of elements constituting a cushioned freight car. Its most significant difference from the "A" and "B" claims however, is its inclusion of "resilient lading" as a positive element of the claimed combination. Furthermore, its definition of the "cushioning means" is somewhat differently phrased — namely, (a) that its closure time is not less than about .23 second at 10 miles per hour, (b) that its stroke does not exceed about 40 inches, and (c) that it has force-travel "relationships" (characteristics) that preclude any sustained lading forces over 2,000 pounds per square foot at 10 miles per hour.
The significance of the stated minimum closure time and maximum travel length or stroke is by now familiar. The allusion to a lading force not exceeding 2,000 pounds per square foot is referable to the disclosure that the species of cartoned canned goods that were used in the impact tests described in the patent were shown by a static test to become subject to damage when the compressive force on them exceeded 2,000 pounds per square foot. Other cartoned canned goods, of course, could have a different damage point.
The term "resilient lading" is not well defined in the patent. While the patent states that such lading is "always characterized by some measure of compressibility," it also declares that "in some instances, it is difficult to categorize a particular lading as either resilient or rigid lading," the latter being defined as lading which has inherent rigidity or substantial non-compressibility. Examples of resilient lading disclosed in the patent include cartoned canned goods, granulated or pulverized material packed in bags or sacks, and, in general, "any form of compressible lading which if over-compacted results in damage to the lading itself or impairs its usefulness."
In summary, when the claims of the Peterson patent are read in the light of the specification and of the relevant parts of the file history, with due regard to the acknowledged fact that a cushion-underframe car using an energy-dissipative type cushion was old, the purported novelty of the Peterson "invention" relates to the particular dissipative cushion employed. Specifically, the alleged innovation is in its travel length or stroke of from 18.6 to 42.8 inches for a cushion having a substantially constant force-travel characteristic, or in the case of a cushion having a lower efficiency, a travel length in a range that is increased above the 18.6 to 42.8 inch range by an amount that will compensate for the reduced efficiency.
While the claims also include a statement as to minimum closure time at a given impact speed, the patent itself discloses that this, in reality, is simply no more than an alternative way of describing the travel length for the 100 per cent efficient cushion having adequate cushioning capacity. Moreover, the trial evidence demonstrates that, for any practical cushion of lower efficiency which meets the test of equivalence in relation to cushion travel, the closure time necessarily will be greater than the minimums stated in the claims. Therefore, the Court finds that the Peterson patent on its face covers an old combination of inter-functioning parts, with purported novelty hinging on a particular dimension or size — the travel length of the energy-dissipative cushion employed.
III. PATENT VALIDITY
Defendant contends that the Peterson "invention" fails to satisfy two fundamental conditions for patentability — the requirements of novelty and invention. Framed in statutory language, the questions posed are: whether 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" ( 35 U.S.C. § 102(b)) (statutory bar); and whether "the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains" ( 35 U.S.C. § 103).
A. Anticipation
In view of the fact that all of the prior art disclosures relied upon by defendant may be fixed in time more than a year prior to the first application for a patent by Peterson, it is unnecessary for the Court to fix a date of conception or reduction to practice. In resolving the questions of patent validity, it is sufficient to recognize that plaintiff claims a conception date as early as June 1, 1955, and his first application for a patent was made on March 5, 1959. It is undisputed that there had been public commercial use of cushion-underframe cars for several decades, and three styles of cushion had been utilized. One of these — the Hydra-Cushion — was of the energy-dissipative type. As already noted, this cushion, embodying the hydraulic resistance principle, had a travel length of 10 inches.
Cushion-underframe cars had been introduced to satisfy a need for greater cushioning capacity. They afforded, at higher speeds, the then acceptable degree of lading protection provided at 3 to 4 miles per hour by draft gears alone. It was a recognized principle that, for a given set of impact conditions, a longer cushion travel length made possible a lower cushion force and, in turn, a lower car body acceleration. Thereby the possibility of lading damage could be, and was, reduced for higher impact speeds.
Plaintiff contends that the subject matter of the patent claims at issue here differs from the freight car cushions found in the prior art with respect to the cushion travel length, closure time and force-travel characteristic. The patent specification and file wrapper demonstrate beyond cavil, however, that the statement about minimum closure time was only an alternative expression — according to a well-known elementary formula — for the 20-inch cushion travel of a 100 per cent efficient model. Moreover, the trial evidence supports the conclusion that any practical cushion of lower efficiency equivalent to a cushion having the allegedly critical travel length would have a closure time in excess of the stated minimum at the given impact speed. Therefore, the physical structure may be described by its travel and force-travel characteristic without reference to its closure time. Furthermore, while the patent states a preference for a substantially constant force-travel characteristic, the specification indicates that wide divergence from the 100 per cent efficient ideal — taken with a corresponding upward adjustment of travel length — nevertheless falls within the claimed invention. Viewed in terms of specific structure, then, the only difference between the claimed invention and the earlier 10-inch travel Hydra-Cushion resides in the feature of travel length. In a review of the prior art references, however, consideration must be given to the force-travel characteristic as well as the travel length in order to establish whether an "equivalent" travel length had been anticipated.
Closure time will again be considered in the course of the opinion in relation to the asserted time-phase or time-lag theory. See p. 303, infra.
As to the specific prior art, defendant relies primarily on five references, several of which are quite old. Two of the earliest publications — the printed publications of Langley (1886) and Schwartz (1913) — described the use of hydraulic cushions as end-of-track buffers to protect railroad vehicles and terminals in the event a train was not brought to a stop before reaching the track-end. The Langley paper discusses the application of a hydraulic cushion which is metered so as to provide a uniform resistive force during the travel stroke. A cushion travel length of four feet (48 inches) is suggested to protect against impacts up to 8 miles per hour, but other instances of cushion travel length, including one of two feet (24 inches), are mentioned. The Schwartz paper described tests of an experimental hydraulic buffer, having a working travel length of 11 feet. A number of impact tests were run, involving varying numbers of cars (1 locomotive to 1 locomotive and 6 cars) and varying impact speeds (2.90 to 8.10 miles per hour), and the resulting attained travel length varied from 2.65 to 6.56 feet. The paper also states that the ideal buffer would have a constant force-travel characteristic.
Contemporary with Langley's publication is the Fenton patent (1899), describing a hydraulic cushion which may be used either as an end-of-track buffer or as a buffer placed on individual railway cars. In its application as an end-of-track buffer, Fenton's invention is said to be capable of stopping a rapidly moving train "within four feet, more or less." The patent fails to make any reference to cushion efficiency.
A much more recent reference is the publication by Fitzjohn (1955) which deals broadly with the subject of high capacity shock absorbers and buffers. His paper describes the construction, merits, and a number of uses for a hydropneumatic buffer which utilizes the hydraulic resistance principle through a metered orifice, supplemented serially by a pneumatic element (as in the accused ACF cushion) which serves to reposition the cushion. A particular form of the hydropneumatic buffer with an over-all travel length of 24 inches is described and schematically drawn; and the same principles are said to be applicable to cushions having travel lengths up to 36 inches.
Sharpest controversy centered around the most recent reference — the article by Myers published in 1957 — which praises the cushion-underframe car for the protection it affords to car and lading, particularly in "high-speed, heavy-impact" situations. The article mentions that certain manufacturer's tests showed that "with a cushioned underframe having a 24-in. yield, the force dealt with amounted to only 137,500 lb." This force level could be "handled by the car and its lading without damage to either." In the immediately following paragraph, the newly developed Hydra-Cushion (10 inch) underframe is briefly described.
Of the numerous objections raised by plaintiff to this particular reference, only some merit consideration. Plaintiff asserts that the reference fails to specify a dissipative type cushion, and fails to discuss the nature of the force-travel characteristic. Moreover, plaintiff argues that the test information referred to was derived from a Pullman-Standard advertisement known throughout the industry; and that the 24-inch yield mentioned was actually the total yield of an impact between two cushion-underframe cars having 8-inch travel rubber cushions and draft gears.
While conceding that the article does not specify the cushion type or force-travel characteristic, defendant asserts that dissipative cushioning was what the art took to be the modern and preferred type, and that the force-travel characteristic becomes immaterial in view of the breadth with which the plaintiff is forced to interpret the patent claims in order to reach the accused cushions. In response to plaintiff's attempt to ally Myer's description of a 24-inch yield to plaintiff's own advertisement for its 8-inch cushion, defendant argues that the origin of a reference is immaterial, and that the patent law gives legal effect to the Myer's article dependent solely on its content.
As the foregoing summary demonstrates, both parties have raised interesting and difficult questions with respect to the individual prior art references. While it might be possible to resolve the focal question of patent validity on the ground that one or more of these references robs the Peterson patent of material novelty, the nature of the proofs strongly suggests that it is unnecessary, and probably imprudent, to assume the arduous task of assessing individual references in order to answer the question of anticipation. In view of the Court's conclusion that Peterson's patent lacks invention because its subject matter was obvious, specific findings on the questions raised by the defense of lack of material novelty will be foregone in favor of a more detailed evaluation of the obviousness of the claimed invention.
B. Obviousness
Only last Term, the Supreme Court had occasion to consider — for the first time since its enactment — that section of the 1952 Patent Act which prescribes the test of nonobviousness as a condition of patentability. In Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966), the primary question posed was whether the 1952 Act had modified the traditional judicially evolved criterion of "invention." The Court concluded that Congress had merely codified the body of precedent which had developed the principles announced long ago in Hotchkiss v. Greenwood, 52 U.S. 248, 13 L.Ed. 683 (1850); the general level of innovation necessary to sustain patentability remained unchanged. In the course of the opinion in Graham, a methodology for evaluating obviousness was described as follows:
July 19, 1952, c. 950, § 1, 66 Stat. 798, 35 U.S.C. § 103 (1964).
See, also, United States v. Adams, 383 U.S. 39, 86 S.Ct. 708, 15 L.Ed.2d 572 (1966), decided the same day.
"Under § 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved. Against this background, the obviousness or nonobviousness of the subject matter is determined. Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc., might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented. As indicia of obviousness or nonobviousness, these inquiries may have relevancy." ( 383 U.S. at 17-18, 86 S.Ct. at 694).
Recently, our own Court of Appeals in Formal Fashions, Inc. v. Braiman Bows, Inc., 369 F.2d 536 (2d Cir. 1966), further clarified the rules laid down in Graham. It characterized the statutory standard as a "specialized reasonable man test for obviousness," observing that proof on this issue "should tend to show what would be obvious to a hypothetical mechanic who, among other things, has the prior art in mind when he endeavors to solve the problem for which the patent is obtained." ( 369 F.2d at 538). See also Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 2 Cir., 372 F.2d 263 (1967).
Having described the applicable methodology, we turn to the more difficult task of application. The state of the art in railroad cushioning and a number of specific prior art references have already been discussed. Several of these references, it should be noted, dealt with cushions designed principally as end-of-track buffers. Whether a cushion so employed is the full equivalent of one placed within the vehicle would be an issue on the defense of anticipation, but there can be no question that these references are pertinent to the test of obviousness. Continental Scale Corp. v. Harrison Wholesale Co., 132 F.2d 463, 465-466 (7th Cir. 1942). Whether the cushion is fixed (as at a track-end or the base of an elevator shaft) or fitted on a moving vehicle (as on an aircraft landing gear strut or the sliding sill of a cushion-underframe car), its function and mode of operation are essentially the same. Indeed, for Peterson's own experimental tests, the friction cushion was mounted on the striking end of the impacting car, while lading forces were measured in the standing car — a conventional, uncushioned freight car.
It is apparent, as our Court of Appeals has observed, that "the more numerous the references and the more remote the cited art from the subject matter of the patent in suit, the less likely it becomes that a person having ordinary skill in the art would have arrived at the result reached by the patent in suit." Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 372 F.2d 263 (2d Cir. 1967).
But the Court need not unnecessarily devote time to an examination of the literal differences between the claimed invention and each of the putatively relevant prior art references, for the question of obviousness can be answered readily and clearly by considering analytically the earlier 10-inch Hydra-Cushion car alone. The parties have stipulated that a dissipative cushion car was in public use before June, 1955; and the evidence reveals that the Hydra-Cushion car was publicly known before the Peterson "invention" was made. We have seen that the sole literal differences between the patent claims and the 10-inch Hydra-Cushion car concern the feature of cushion travel (and corresponding closure time for a given speed), and — in relation to the "B" type claims — the element of cushion efficiency
Under § 103, the subject matter of the patent is to be tested against the prior art and the relevant learning as of "the time the invention was made." While this "time" is not necessarily the patent application date, Application of Palmquist, 319 F.2d 547, 51 CCPA 839 (1963), the Court of Customs Patent Appeals recently held that the proponent of a patent cannot dispose of references which antedate the patent application by more than a year through proof of an earlier invention date. Application of Foster, 343 F.2d 980 (C.C.P.A. 1965), cert. denied, 383 U.S. 966, 86 S.Ct. 1270, 16 L.Ed.2d 307 (1966); Application of Folkers, 344 F.2d 967 (C.C.P.A. 1965). Cf., Hazeltine Research, Inc. v. Brenner, 382 U.S. 252, 86 S.Ct. 335, 15 L.Ed.2d 304 (1965). However, the question of a time bar under § 103 is academic here, because there is no basis in the record for a finding that the alleged invention was made before the Hydra-Cushion was publicly known.
The specification of cushion efficiency can be dealt with summarily. Plaintiff has asserted the patent claims — including the "B" type claims — against the 20-inch Hydra-Cushion, which the evidence shows was of comparable efficiency to that of the 10-inch travel model. And so, to make out a case of infringement, plaintiff has sought to establish an exceedingly liberal gloss on the efficiency specification. Indeed, if the Court were to accept the plaintiff's construction, any practical cushion would appear to meet the proffered standard. We might say simply: plaintiff cannot have its cake and eat it too. But interpretation of patent claims is not left to the discretion of the patent owner; it is a matter for the Court.
While the patent states a preference for high efficiency, it also reveals that high efficiency is not critical to the claimed invention. What is critical, according to the patent specification, is a combination of cushion travel length and cushion efficiency. For the preferred 100 per cent efficient cushion, a range of about 20 to 40 inches is specified. Yet lower efficiencies seem very much acceptable, provided they are accompanied by longer travels, furnishing the same minimum closure time and equivalent energy absorption with the same maximum cushion force. Given this interpretation, the arrangement comports with the laws of mechanics insofar as they are pertinent to the cushioning phenomenon.
When the patent states time and efficiency qualifications are thus disposed of, it becomes manifest that the only material physical distinction between the claimed invention and the earlier Hydra-Cushion consists of the difference in the length of the cushion travel. Consequently, although plaintiff advances certain allegedly "unknown" engineering principles concerning the force and time relationships involved in lading protection, the truly critical question of patentability is whether it would have been obvious — in light of the prior art and the "known" principles — to utilize cushions of greater travel length to achieve better lading protection. Mackay Radio Telegraph Co. v. Radio Corp. of America, 306 U.S. 86, 94, 618, 59 S.Ct. 427, 83 L.Ed. 506 (1939).
See pp. 293, 294 supra.
To this question the trial evidence most convincingly supplies an affirmative answer. Although the Court has deemed it unnecessary to adjudicate the issue of anticipation, the prior art references considered individually and as a group demonstrate a broad and understanding awareness of the possibilities for cushion travel length in the allegedly critical range. Apart from the disputed Myer's article (with its explicit allusion to a "cushioned underframe having a 24 inch yield"), there are a number of other references that describe comparable travel lengths in cushions employed in analogous apparatus, including end-of-track buffers, elevator shafts, and aircraft landing gear struts.
Moreover, even if it were assumed that the use of a freight-car cushion of the specified length, or its equivalent, is not made obvious by the referenced patents and publications, the fully known mechanics of cushion operation would have, and did, compel consideration of longer travels. We have already pointed out that it is inherent in the operation of a cushion that longer travel may be utilized to increase cushioning capacity (and thereby to afford lading protection at higher speeds) or to reduce maximum cushion force (and thereby to afford more comprehensive lading protection at a given speed).
It is immaterial whether the Court adopts plaintiff's or defendant's description of the "man having ordinary skill in the art." Manifestly, this "hypothetical mechanic" would be familiar with the principles of cushioning and with the application of cushioning devices to the protection of car and lading.
Plaintiff does not dispute that increased cushioning capacity and/or lower cushion force were known benefits of longer cushion travel. However, plaintiff does contend that, nevertheless, it would not have been obvious to go to longer travels in freight car cushions because the anticipated benefits, in terms of lading protection, would have been negligible. Otherwise expressed, plaintiff's contention is that, while greater travel length may have been obvious in the sense that it was obviously available, no one having ordinary skill in the art would have chosen to go to the range specified in the patent to achieve lading protection.
The logic of this proposition is refuted by another of plaintiff's own contentions. A basic premise of the patent and of plaintiff's case for validity is that, before Peterson, those skilled in the art believed lading force and cushion force to be roughly proportional. The patent itself indicates that an extension of travel from 10 to 20 inches will yield a 50 per cent reduction in cushion force. Even on plaintiff's premise, then, it would have been "known" that lading forces could be halved by increasing cushion travel from 10 to 20 inches. With that point conceded, such an extension in travel seems "obvious" as a matter of law. In any event, it is obvious as a matter of established fact.
Plaintiff suggests that, nevertheless, a proportional reduction in lading force would have been considered futile, because the absolute magnitude of the lading forces, in a 10-mile-per-hour impact, still would have been above the damage level for the allegedly critical, "resilient" class of lading. This point can be disposed of without even considering the qualification as to speed and lading. The suggestion flies in the face of still another of plaintiff's contentions — namely, that before Peterson no one had attempted to solve the lading protection problem by making direct lading force measurements.
Plaintiff's position, amply supported by the evidence, is that the common approach had been to design a cushion which, at higher impact speeds, provided the same maximum cushion force obtained with standard draft gears alone at the lower impact speed of 4 miles per hour. With cushion or coupler force comparable, lading force also would be comparable; and a satisfactory level of lading protection would be achieved. How then, one may ask, would those working on the problem have been discouraged from going to longer travel lengths if they had not concerned themselves with the specific magnitude of the lading forces?
The question is rhetorical. Its answer is as obvious as the use of greater travel length to achieve better lading protection. If the matter is as simple as has just been suggested, it becomes a logical next point of inquiry for the Court to ascertain the basis upon which Peterson obtained his patent. Before the Patent Office and at the trial, plaintiff claimed certain previously unexpected results which were said to occur when cushion travel was extended into the specified range. These assertedly unanticipated and surprising results pertain to two allegedly unknown relationships: the force disproportionality and the lading compaction-cushion closure time lag. Both concepts merit discussion by the Court.
The Force Disproportionality
What is called the force disproportionality is a difference that Peterson found to exist between (a) the percentage reduction in lading force and (b) the percentage reduction in coupler or cushion force, on a comparison of his test results with two different cushion travel lengths. Stated simply, it was found that, as cushion travel length was increased, lading force was reduced by a larger percentage than the percentage of reduction in cushion force. This relation is called a "disproportion" in contrast to the postulate ascribed by the patent to the prior art, that lading force and cushion force would be expected to be reduced by roughly the same percentage. Thus, plaintiff's patent itself recognized as prior knowledge that both a lower lading force and a lower cushion force can be effectuated by lengthening cushion travel. The disproportion is simply a finding concerning the allegedly unexpected magnitude of the known reduction of lading force in comparison to what was to have been anticipated based on the postulated prior art.
In view of the presumption of validity ordinarily attaching to a duly issued patent, it is particularly noteworthy that, while the patent disclosure and file wrapper suggest a criticality for the specified cushion travel range in regard to the asserted disproportion, the facts appear otherwise. As Peterson conceded upon his cross-examination at trial, the disproportion is not peculiar to the 20- to 40-inch range, or its equivalent, but attends differences in cushion travel below that range. Even on plaintiff's postulated prior art, then, the asserted disproportion is not something unique to the specified apparatus. It is a relationship which, to a varying degree, attends any increase in freight car cushion travel. At best, Peterson could claim only to have gained pioneering data concerning this inherent effect of increasing cushion travel, unknown before because no one previously had made concurrent measurements of lading force and cushion force. The disproportion remains only a quantitative aspect of a known effect, and, as already noted, is not even peculiar to the travel range specified in the patent.
35 U.S.C. § 282 (1964). See, also, Mumm v. Jacob E. Decker Sons, 301 U.S. 168, 57 S.Ct. 675, 81 L.Ed. 983 (1937); Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 372 F.2d 263 (2d Cir. 1967); Buxton Inc. v. Julen Inc., 223 F. Supp. 697 (S.D.N.Y. 1963). But see Graham v. John Deere Co., 383 U.S. 1, 18-19, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966); Report of the President's Commission on the Patent System, pp. 22-23 (1966); Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., supra 372 F.2d at 270 (dissenting opinion of Kaufman, J.).
"A patentee may not arbitrarily select a point in a progressive change and maintain a patent monopoly for all operations in that progressive change falling on one particular side of that arbitrarily selected point. It is only where the selected point corresponds with the physical phenomenon and the patentee has discovered the point at which that physical phenomenon occurs that the maintenance of a patent monopoly is admissible. A claim must be based on invention. The claim to invention depends here upon an alleged discovery of certain limits or points which do not exist in fact and there is therefore no invention."
Kwik Set, Inc. v. Welch Grape Juice Co., 86 F.2d 945, 947 (2d Cir. 1936).
"The mere location of the optimum conditions of use for a known composition of matter does not constitute `invention' so as to entitle the discoverer thereof to a monopoly. That objective, however useful the final result, can be achieved by `patient experiment'."
Helene Curtis Industries, Inc. v. Sales Affiliates, Inc., 233 F.2d 148, 152 (2d Cir. 1956), cert. denied, 352 U.S. 879, 77 S.Ct. 101, 1 L.Ed.2d 80 (1956) (patent for hairwaving chemical composition invalidated).
See also Barber-Coleman Co. v. A.G. Redmond Co., 94 F.2d 717 (6th Cir. 1938), where the Sixth Circuit Court of Appeals said, in invalidating a patent for an induction motor:
"The present inventors may have found by trial and error the precise [proportions] that would give the best result, but they discovered no new principle, disclosed no new combination of elements, and, if they secured a better result, it was an improvement but in degree, and for this the law does not award a patent, however persistent and bold in research expenditure the applicants therefor may have been." 94 F.2d at 720.
It is not without significance that the impact tests which provided the data revealing the disproportion were performed solely with the experimental straight friction cushion. In later tests of plaintiff's own 20- and 30-inch travel hydraulic cushions, the disproportion in the reduction of forces with longer travel was substantially less than was found with the straight friction cushion. The published report of these later tests showed a significantly closer correlation between the reduction of lading force and coupler force, particularly with lighter loads. Peterson himself made no mention of an unexpected disproportion in his 1958 management report on the experimental friction cushion tests, or in any writing prior to his second patent application of December 3, 1959. In fact, the 1958 report concluded that the test results were in "good general agreement * * * with basic dynamic theory."
Quite to the contrary, his report recognized that the possibility of a disproportion would not be surprising to those skilled in the art. See quoted passage at p. 303, infra.
It appears manifest from the pertinent authorities that Peterson's experimental findings cannot lift the otherwise obvious subject matter of his patent to the status or quality of an "invention." We start with the general rule, as formulated nearly a century ago in Smith v. Nichols, 88 U.S. 112, 119, 22 L.Ed. 566 (1874):
"a mere carrying forward or new or more extended application of the original thought, a change only in form, proportion 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."
While the recent decision of the Supreme Court in Graham v. John Deere Co., supra, outlines a specific methodology for evaluating the obviousness of an alleged invention, it explicitly found that the statutory test had not been intended to change the general level of patentable invention. Consequently, the body of judicial precedent which has developed and refined the principles stated in Smith v. Nichols still has relevance in appraising the obviousness of a purported invention.
It has become a familiar rubric that a mere change in degree will not give rise to a patentable invention. Putting aside for the moment the asserted disproportion, we regard the subject matter of the Peterson patent as falling squarely within this category of degree. The extension of travel and the preference for high cushion efficiency represent, at best, a "mere carrying forward * * * or more extended application of the original thought * * *." It is an axiom of cushioning principles and experience that an extension of travel will provide in greater degree what cushioning has always provided. The preference for high efficiency serves only to achieve these benefits with minimum travel.
See Walker, Patents § 116 n. 1 and accompanying text (Deller, 2d ed 1964). It also is said that to change the proportions of an old combination will seldom amount to invention. (Walker, supra, § 115.) While an exception was created when the change was critical as compared to the proportions of the prior art, In re Waite, 168 F.2d 104, 35 CCPA 1117 (1948), a patent will not be granted for the mere discovery by routine experimentation of an optimum range. In re Aller, 220 F.2d 454, 42 CCPA 824 (1955); Helene Curtis Industries, Inc. v. Sales Affiliates, Inc., 233 F.2d 148 (2d Cir.), cert. denied, 352 U.S. 879, 77 S.Ct. 101, 1 L.Ed.2d 80 (1956); Kwik Set, Inc. v. Welch Grape Juice, 86 F.2d 945 (2d Cir. 1936).
The Court's judgment — shaped by the persuasive evidence — is that Peterson's patent involves "a matter of degree rather than a change in kind" in the sense that his appreciation of the significance of lengthening the cushion stroke to the range of about 20 inches to 40 inches "did not require the level of intellectual effort and perception which entitles" his device — whose novelty consists solely of a longer cushion stroke — "to statutory protection." Georgia-Pacific Corp. v. United States Plywood Corp., 258 F.2d 124, 132 (2d Cir.), cert. denied, 358 U.S. 884, 79 S.Ct. 124, 3 L.Ed.2d 112 (1958).
See pp. 284-285 supra.
Carved from the general rule, that a change in degree does not amount to invention, is the narrow but significant exception first articulated by the Supreme Court in Eibel Process Co. v. Minnesota Ontario Paper Co., 261 U.S. 45, 43 S.Ct. 322, 67 L.Ed. 523 (1923). In that case, the subject matter of the patent involved a mere change in the degree of pitch of the wire-mesh belt used to form the paper on a Fourdrinier paper-making machine. The productive capacity of such a machine is proportional to the speed of the belt. Before Eibel, this speed had been limited because, above a certain level, ripples were produced which impaired the quality of the paper. By giving the belt a substantial positive pitch, Eibel was able to take advantage of gravity to supplement the frictional drag of the wire in moving the pulp. Previously, the belt had been given a small positive or negative pitch, depending upon the character of the wet pulp, to control the extent of water drainage as the pulp moved between the receiving point and the first roll. While Eibel's invention gave no further improvement in drainage, it solved the problem of rippling, thereby permitting much higher belt speeds. This was not a mere extension of the original thought about pitch, but brought in a new mode of operation to achieve a completely different purpose.
Indeed, Eibel's invention actually impaired drainage for those kinds of pulp which called for a negative pitch.
In sharp contrast, the extension of cushion travel — the foundation of the Peterson patent — involves neither a new action nor a different objective. Plaintiff has conceded that the asserted disproportion is not uniquely present in the specified range of cushion travel, but attends extension of travel below the assertedly critical range. Moreover, this increase in travel length serves the identical purposes of prior increases in travel length — i.e., the reduction of lading forces and the enlargement of cushion capacity.
The facts in the case at bar are less persuasive for sustaining the patent than those in General Electric Co. v. Cooper-Hewitt Electric Co., 249 F. 69 (6th Cir. 1918), in which the patent was invalidated. In that case, the patentee had uncovered a theretofore unknown effect in a particular range of input values. Nevertheless, patentability was denied, because the specified range was one to which the art was free to go in order to achieve in higher degree another effect which was already known. By contrast, the "disproportion" asserted here involves the same effect of lowered lading forces from increased cushion travel that was known to the prior art.
Plaintiff's reliance on the recent decision in United States v. Adams, 383 U.S. 39, 86 S.Ct. 708, 15 L.Ed.2d 572 (1966), is based on a misunderstanding of the facts involved in that case. There, the patentee had incorporated known elements in a new combination having unexpected operating characteristics. Two "long-accepted" technical considerations were found to have previously deterred investigation into such a combination.
On the other hand, in the present litigation, the combination was old and the alleged invention turns on a mere specification of certain characteristics. Accepting, arguendo, plaintiff's contention that those skilled in the art assumed coupler and lading force to be proportional, the use of a cushion falling within the patent claims was quite obvious. The factors which operated to "deter" such use, if that is a proper characterization, are not to be found in the "known" technology of cushioning at the time of the alleged invention.
The decision by our Court of Appeals in Edison Electric Light Co. v. United States Electric Lighting Co., 52 F. 300 (2d Cir. 1892), likewise is inapposite. Apart from the several differences in physical structure between the patented lamp and the prior art, controlling effect was given to the finding that the complete combination "for the first time in the art produced a practical electric light."
The facts and circumstances in Goodyear Tire Rubber Co. v. Ray-O-Vac Co., 321 U.S. 275, 64 S.Ct. 593, 88 L.Ed. 721 (1944); Minerals Separation, Ltd., v. Hyde, 242 U.S. 261, 37 S.Ct. 82, 61 L.Ed. 286 (1916); Gasoline Products Co. v. Coe, 66 App.D.C. 333, 87 F.2d 550 (1936); Georgia-Pacific Corp. v. United States Plywood Corp., 258 F.2d 124 (2d Cir.), cert. denied, 358 U.S. 884, 79 S.Ct. 124, 3 L.Ed.2d 112 (1958); Dewey Almy Chemical Co. v. Mimex Co., 124 F.2d 986 (2d Cir. 1942); Refractolite Corp. v. Prismo Holding Corp., 117 F.2d 806 (2d Cir. 1941); Toledo Computing Scale Co. v. Computing Scale Co., 208 F. 410 (7th Cir. 1913); Gulf Smokeless Coal Co. v. Sutton, Steele Steele, 35 F.2d 433 (4th Cir. 1929), cert. denied, 280 U.S. 609, 50 S.Ct. 158, 74 L.Ed. 652 (1930); and in the numerous other cases cited to the Court by plaintiff are so dissimilar from those in the case at bar as not to warrant individual discussion.
In summary, a careful review of the pertinent authorities confirms the Court's finding and conclusion that the asserted disproportion fails to invest the claimed apparatus with the character of an invention. Nothing in the recent opinions in Formal Fashions, Inc. v. Braiman Bows, Inc., 369 F.2d 536 (2d Cir. 1966), and Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 372 F.2d 263 (2d Cir. 1967), suggests a contrary conclusion.
The foregoing reasoning has rendered it immaterial whether the disproportion itself would have been obvious to our hypothetical mechanic. Nevertheless, in the interest of completeness, the Court will take up this much-disputed question. As has been observed repeatedly, the patent works from the premise that prior art thinking viewed lading force and cushion force to be roughly proportional. Yet, plaintiff has shown that the common approach to freight car cushioning before Peterson had not concerned itself specifically with lading forces. Rather, those dealing with the problem had adopted a less rigorous protection criterion by seeking merely to maintain the coupler force, at higher impact speeds, at a level comparable to that prevailing with standard draft gears alone in a 4-mile-per-hour impact.
While the testimony of the respective expert witnesses of the opposing parties is in conflict, the far greater weight of the credible evidence supports the conclusion (urged by defendant) that those having ordinary skill in the art would not have thought the possibility of a disproportion to be surprising. We have already had occasion in the course of this opinion to point out that it is virtually impossible to make an accurate theoretical calculation of the relationship between lading force and coupler force. Nonetheless, certain aspects of the problem would be self-evident even to a person unskilled in the art or lacking more general engineering expertise: If an object rests within a car, and the car is accelerated by some outside force, the object must somehow be given the same acceleration if it is to move with the car. Friction between the object and the car provides a limited force which tends to make the object move with the car. However, if the car acceleration is too abrupt, the limited friction force is insufficient to give the object the same acceleration. Some vertical support — such as the car endwall — must come into play to force the object to move with the car.
An undergraduate engineering student, asked to draw a simple force diagram of the problem, would be expected to recognize the existence of the friction force, and would know of its relative constancy. Asked to analyze the relationship between the external force and the endwall force on the object, he would be able to predict some disproportionality based on the relative constancy of the frictional force. A fortiori, a competent engineer of the prior art adopting Peterson's approach would have perceived the possibility of a disproportion.
Recognition of the role of friction in lading protection is confirmed by several prior art writings, including a patent for anti-skid abrasive paint for use on car floors. Further confirmation comes from Peterson's own pen. His 1958 report to management states: "it is well known in engineering dynamics that when two masses are resiliently connected to each other so that relative motion can occur between them, such as a load shifting in a car body, a shock load applied to one is not necessarily proportional to the shock load experienced by the other."
In a final rebuttal on the question of the obviousness of the disproportion itself, plaintiff argues that a competent engineer of the prior art also would have noted the vertical bounce of freight cars upon impact, recognized the consequent loss of frictional contact, and therefore concluded that friction was an insignificant factor in accelerating the lading. Yet this argument overlooks another known effect of increasing cushion travel: the reduction of this vertical motion which disrupts the frictional contact. It is, therefore, unreasonable to assume that those skilled in the art would have totally discounted the effect of friction in moving the lading.
Lading Compaction-Cushion Closure Time Lag
The other allegedly significant but previously unknown relationship relied upon by the plaintiff concerns the time phase or lag between (a) the instant when the force on the lading has reached its peak ("maximum compaction") and (b) the instant when the cushion has fully closed ("cushion closure"). In support of the patent, plaintiff asserts that the claimed apparatus permits maximum compaction to precede cushion closure, a phase relationship which is said to afford protection to the lading. In presenting this time phase theory, plaintiff seems to suggest that each kind of lading has an inherent compaction time, or at least an inherent range of compaction times. Peterson is alleged to have discovered that, by utilizing a cushion of the specified travel length and efficiency or its equivalent, the previously unknown desired time lag will be obtained. The contentions are entirely misleading.
To begin with, there is no inherent compaction time for a given lading. Compaction time is a function of the force on the lading. The force, in turn, is as much dependent on the character and mode of operation of the entire system, including the cushion itself, as it is upon the characteristics of a particular kind of lading. Second, the closure time and the desired time phase or lag is not peculiar to cushion travels in the 20- to 40-inch range. As the patent specification reveals, closure time is dependent on impact speed (among other variables); and the desired time phase occurs with shorter cushion travels at impact speeds within the cushion's capacity. There is nothing critical about the 10-mile-per-hour speed which was used as a standard of reference in the patent. It has significance only insofar as it represents the patentee's judgment in regard to the level of protection which ought to be provided.
Under close scrutiny, the overblown time phase theory proves to be no more than an alternative expression for the known design principle that the energy absorption capacity of a cushion should be sufficient to handle the impact energy received. The case properly contrasted with the desired time phase is one in which the cushion has been overtaxed. In such a situation, the cushion has closed before the impact energy is spent, and the remaining energy is transmitted to the car as an uncushioned blow. This causes an abrupt further acceleration of the car and a resulting higher, potentially damaging, force on the lading. It was well known in the prior art, however, that greater cushioning capacity may be achieved with longer cushion travel, and that a properly designed cushion does not leave unspent energy to be transmitted as an uncushioned blow. In this respect, therefore, the Peterson patent adds nothing materially significant to the sum of knowledge concerning correct cushioning, but merely dresses an old and basic idea in new clothes. In view of Peterson's testimony on this matter, and the testimony of plaintiff's own expert, it was somewhat surprising to find plaintiff still relying on this theory in its post-trial briefs.
One final observation about the time phase relationship deserves mention. By itself, achieving the desired time phase is not a satisfactory criterion for lading protection, for it is not a direct indication of either the cushion force or the resulting lading forces. Maximum compaction can occur prior to cushion closure, yet there can be damage to the lading if that peak lading force is high enough; and, conversely, the peak lading force can be reached after cushion closure and still not cause damage, if its magnitude is below the damage level for the given lading. The patent's correlation of cushion closure time and maximum compaction time, therefore, is not one which reasonably assures lading protection.
Secondary Considerations Bearing on Obviousness
Inquiry into the prior art and the level of ordinary skill in the field has shown with clarity the technological obviousness of the subject matter of Peterson's patent. Judicial precedent, however, has long compelled examination of certain additional circumstantial matters — such as long felt but unsolved need, failure of others, commercial success — in evaluating the inventive quality of a patented innovation. In the recent decision in Graham v. John Deere Co., supra, these considerations were termed "secondary," although it was noted that they "may have relevancy" as "indicia of obviousness." ( 383 U.S. at 18, 86 S.Ct. 684).
See, e.g., Goodyear Tire Rubber Co. v. Ray-O-Vac Co., 321 U.S. 275, 64 S.Ct. 593, 88 L.Ed. 721 (1944); Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 372 F.2d 263 (2d Cir. 1967); Edison Electric Light Co. v. United States Electric Lighting Co., 52 F. 300 (2d Cir. 1892). See, generally, Comment, "Subtests of `Nonobviousness': A Nontechnical Approach to Patent Validity," 112 U. Pa.L.Rev. 1169, 1172 (1964).
The Court observed that these subtests focus attention on "economic and motivational" issues which are "more susceptible of judicial treatment than are the highly technical facts often present in patent litigation." Moreover, they serve to "guard against slipping into use of hindsight" in applying the statutory test. Nevertheless, the Court suggested that these desiderata only rarely can "tip the scales of patentability" when the technical inquiry leaves little doubt. Cf. Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 372 F.2d 263 (2d Cir. 1967).
383 U.S. at 36, quoting from Monroe Auto Equipment Co. v. Heckethorn Manufacturing Supply Co., Inc., 332 F.2d 406, 412 (6th Cir.), cert. denied, 379 U.S. 888, 85 S.Ct. 160, 13 L.Ed.2d 93 (1964). See also Diamond Rubber Co. v. Consolidated Rubber Tire Co., 220 U.S. 428, 31 S.Ct. 444, 55 L.Ed. 527 (1911); Preformed Line Products Co. v. Fanner Mfg. Co., 328 F.2d 265, 273 (6th Cir.), cert. denied, 379 U.S. 846, 85 S.Ct. 56, 13 L.Ed.2d 51 (1964); Elgen Mfg. Corp. v. Ventfabrics, Inc., 207 F. Supp. 240, 251 (N.D.Ill. 1962), aff'd, 314 F.2d 440 (7th Cir. 1963).
To a considerable degree, plaintiff has avoided confrontation with the principal technical issues in its overly argumentative post-trial submission of proposed findings of fact and conclusions of law. Instead, plaintiff posits major emphasis on a number of these secondary factors. In view of plaintiff's heavy reliance on these matters, and for the purpose of pointing out why long-travel cushions had not earlier found favor in the industry, the Court shall now consider briefly the history of freight car cushioning in relation to the lading damage problem.
Proof of a long-felt demand has probative value under the statutory test because it is reasonable to infer that the need would be filled if the solution were obvious. The parties have stipulated that "prior to Peterson's alleged invention, there had been a long recognition, expressed by many, that there ought to be more done about protecting lading. The damage losses and cost to the railroads were high. Many measures were taken, including cushioning." (R. 451-52). Supplementing this stipulation, plaintiff introduced into evidence a number of published articles which referred to the lading damage problem and the need for a solution.
See Comment, 112 U.Pa.L.Rev. 1169, 1172 (1964).
The cited publications also served to point up the diversity of emphasis of those concerned with the problem. There was no suggestion of any lack of know-how about cushioning. Moreover, in response to plaintiff's proof on this point, defendant introduced several writings indicating that the lading damage problem has persisted after the Peterson patent. In fact, damage has been reported even on plaintiff's own hydraulic cushion-underframe cars having 30 inches of cushion travel.
The inferential significance of a showing of long-felt need may depend on numerous desiderata, among them, the economic cost sustained through lack of a solution of the problem, the general responsiveness of the industry, and the quality and quantity of research being conducted in the field. While plaintiff would have this court characterize the prior art cushions as "unsatisfactory," the evidence shows that earlier freight car cushions, like the more recent longer travel cushions, afford additional protection against damage under impact conditions within their respective capacities. In sum, the totality of proof on the circumstances of long-felt demand does not provide a basis for a meaningful inference in regard to the nonobviousness of Peterson's alleged invention.
At the trial, plaintiff devoted considerable time to proving, in detail, the course of Peterson's activity from early 1955 through 1959. Regardless of the purpose of such proof, the evidence clearly gives little support to plaintiff's contention of nonobviousness. The record shows no more than a relatively thorough but routine approach to the question of how much cushion travel is enough to satisfy someone's judgment about the level of protection which ought to be provided.
The story begins in late 1954 when Peterson, an engineer in plaintiff's Research and Development Department, was assigned to the subject of freight car cushioning. Peterson informed himself on the subject by talks with his predecessor and by a review of plaintiff's research files, but without a search of other publicly available literature. On the basis of some simple engineering calculations, Peterson concluded that a cushion having 11 inches of travel would be insufficient, at 10 miles per hour, to protect cartoned bottle goods. At that time, he urged that direct tests be made to learn just what cushion travel would be needed, and to learn what could be done in regard to lading protection if a limitation of 11 inches of cushion travel were imposed.
Authorization was received from plaintiff in 1966. Tests were run in the latter half of 1957 under Peterson's supervision. This test program furnished limited numerical data concerning the force levels and time relationships with combinations of different travels and varying extents of freight car compartmentalization. No mention of the particular range of cushion travel from 20 to 40 inches, or of the other cushion characteristics stated in the Peterson patent, may be found in his management report of 1958 or in any other writing produced by him before the patent application of March, 1959. The simple conclusion called for by the proof is that Peterson conducted a straight-forward investigation of force and time factors over a range of cushion travels and degrees of compartmentalization. His purpose was to gain specific information on cushion design and compartmentalization criteria for the protection of lading.
See p. 289 supra.
As part of its circumstantial case for nonobviousness, plaintiff also has sought to prove that the growing interest in longer travel cushions, which began in the late 1950's, was generated by disclosures of its own developmental work in the field. On the other hand, defendant has attempted to establish instances in which others in the art came independently to the subject matter of the Peterson patent at about the same time as Peterson. While the proof as to a number of specific factual issues is conflicting, the fair preponderance of the credible evidence requires the Court's finding that others in the pertinent art were engaged in independent development of longer travel cushions for use in freight cars. To the extent that this finding provides some basis for an inference that the "solution" was already known, it has some probative value in confirmation of the obviousness of the alleged invention.
In particular, plaintiff charged deliberate copying on the part of the defendant, relying primarily on an alleged communication to one of defendant's leading engineers in early October, 1958, and a "spying" incident in May, 1958. Whatever bearing this accusation may have on the question of obviousness, "derivation" has not been established; and there is a total failure of proof on "identity" apart from the common cushion travel length.
See Comment, 112 U.Pa.L.Rev. 1169, 1180 (1964) ("simultaneous solution").
The Court also has been asked to consider the commercial success of cushions alleged to have the characteristics called for in the Peterson patent. The rationale underlying this subtest is that the likelihood of commercial gain would have induced earlier production of the "invention" if its subject matter were obvious. It is undisputed that there has been an enormous growth in the production of cushion-underframe cars. Although a number of different cushion travel lengths have been utilized, recent sales have been concentrated at lengths of 10, 20 and 30 inches. It is axiomatic, of course, that at the same efficiency, a cushion of longer travel is capable of providing better lading protection. However, while sales of 20-inch cushions have dwarfed the sales of all other cushions, sales of 10-inch cushions have greatly exceeded those of 30-inch models. The over-all sales picture portrays not only a surge of interest in cushioning but also a continuing division of opinion in the railroad world as to how much cushioning is enough.
Id. at 1175.
For example, in the 10-month period ending on June 30, 1965, sales of 10-, 20- and 30-inch cushions were as follows: 10-inch, 5,477; 20-inch, 21,069; 30-inch, 1,565.
Notwithstanding the foregoing remarks, the impressive recent commercial success of longer travel cushions cannot be ignored. In light of the dual circumstances of commercial success and apparent technological obviousness of the alleged invention, there is strong reason to inquire why longer travel cushions did not gain favor in the industry earlier. The answer is not singular and well-defined. A complex of factors seem to have been operating to deter or dissuade the railroad industry from earlier use of longer travel cushions.
The Court has noted that the lading damage problem had been causing concern in the railroad industry for many years. In general, the railroads had responded by attempting to control impact speeds, by urging more care in the packaging of lading, and by using compartmentalizers and other means of bracing lading in cars. Comparatively little interest was shown in the cushion-underframe car. Although the Duryea car — introduced in the late 1920's — had gained some acceptance, it became discredited by structural deficiencies in its sliding sill construction which caused excessive maintenance problems. The higher cost of producing cushion-underframe cars was another important factor.
Others opposed the cushion-underframe design for the reason that problems were created by the extension of the couplers beyond each end of the car. Because prevailing regulations limited the extension of the running-boards atop freight cars beyond the car ends, the increased distance between cars meant a larger gap between the running-boards of adjacent cars. The extension of the couplers also created what was regarded by some as a hazard in coupling and uncoupling. It, therefore, required a departure from standard couplers and air brake connections. Moreover, it created a problem at sharp curves, and at loading platforms where the car length extension prevented registry of car doors with warehouse doors spaced to match conventional freight cars.
A number of these factors tended not only to inhibit widespread adoption of the cushion-underframe design but also to deter development of longer travel cushions even after the cushion-underframe had gained acceptance in the industry.
Several railroad employee unions registered formal opposition with the I.C.C. in 1961, seeking to bar use of plaintiff's 30-inch travel cushion-underframe car. Their joint petition, based on various provisions of the Safety Appliances Acts, was eventually denied after a full hearing.
With plaintiff's introduction in 1951 of a rubber cushion of 7, and then 8, inches of travel, followed by the development of the 10-inch Hydra-Cushion a few years later, there was a renewed interest in the use of improved cushioning as a means of lading protection. The latter device in particular, confirming the known superiority of hydraulic cushioning, allayed the fears of many that hydraulic units would be infeasible for railroad use because of a need for burdensome maintenance. Considering the nature of the railroad industry as a whole, the Court is of the opinion that only a relatively brief period of time passed between the development of the 10-inch Hydra-Cushion and the introduction of longer travel cushions.
In summary, the reluctance of the railroad industry to employ cushions of 20- to 40-inch travel is not attributable to an absence of cushion technology. Instead, the answer is found in the circumstances of railroad operation and railroad economics, deficiencies in the technology of sliding sill construction, and traditional habits and attitudes of mind, well founded or not, that affected the willingness to adopt and use the known cushion technology.
IV. SECTION 112 DEFENSE
The defendant has conceded in its post-trial brief that the defense of invalidity based on a failure to comply with the specificity requirement of Section 112 ( 35 U.S.C. § 112) would have merit only if the patent were given the broad interpretation sought by the plaintiff. Having read the patent claims in the light of the entire specification and the file wrapper history, the Court has given the patent a construction substantially similar to that sought by the defendant. Accordingly, the Section 112 defense is rejected.
V. INFRINGEMENT
Ordinarily, this Court would not refrain from making complete, detailed findings on all of the infringement issues merely because it had declared the patent invalid. It is recognized that the "approved procedure" is to decide the "question of alleged infringement." Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., supra 372 F.2d at 265. A comprehensive study of the evidence, however, has led the Court to conclude that there are critical gaps and other deficiencies in plaintiff's proof of infringement for each of the five accused devices.
For example, plaintiff has introduced no evidence to prove the manufacture, use or sale by defendant of cars employing the early Bendix 30-inch cushion. Plaintiff has failed to show that the force-travel data relied upon against the Hydra-Cushion or Keystone cushions was based on tests of cushions actually used in defendant's cars. Nor did plaintiff succeed in demonstrating that that data was representative or typical of the performance of cushions passing through defendant's hands. For none of the devices, except the Hydra-Cushion (PX-135, 162), was any data introduced which showed the variation in force-travel characteristic over a range of impact mass conditions.
With respect to each of the three accused cushions having a travel length of 20 inches, the available efficiency data for the test impacts and the shape of the force-travel curves (PX-105, 107, 135, 145, 162, 163, 188) indicate that the respective efficiencies of the particular test cushions are so considerably below 100 per cent that plaintiff has even failed to show that any of these particular examples met the characteristics specified in the "A" or "B" type claims. Moreover, on the single "C" type claim, asserted against only the Hydra-Cushion, plaintiff failed to prove that any Hydro-Cushion car sold by defendant had the force-travel characteristic on which plaintiff relies, or that any such car was used to carry a resilient lading and receive the claimed level of protection.
Plaintiff likewise failed to meet its burden of proof under the "B" type claims in regard to the efficiency of the two accused 30-inch travel length cushions (PX-101, 107, 164, 165). Furthermore, because the patent fails to disclose how to compute the efficiency of a hydraulic cushion (whose force-travel characteristic varies under different impact mass conditions) for the purpose of comparing that cushion with the reference cushion of the "A" type claims, it is not possible to determine whether the accused 30-inch travel cushions are equivalent — in the sense of the patent — to a 100 per cent efficient cushion having a travel length of 18.6 inches or higher.
The record does reveal that — in contrast to the preference stated in the patent — all five of the accused devices achieve a gradual buildup of cushion force over a significant portion of the nominal travel length. Finally, in light of plaintiff's failure to prove that any of the accused devices meets the specified characteristics of the patent claims, it is noteworthy that no evidence was introduced to show whether any of the accused devices obtains the allegedly unknown or unexpected results asserted in the patent specification.
The absence of proof on this point is particularly significant in view of the considerably smaller disproportion between lading force and cushion force of plaintiff's own commercial hydraulic cushions in comparison with the straight friction cushion employed in Peterson's initial experimental tests.
The fatal failure of proof on the question of infringement may be at least partially attributable to plaintiff's preoccupation at trial with defendant's strong challenge to patent validity. Moreover, the Court's construction of the patent claims, as heretofore described, differs from that asserted by the plaintiff. In an attempt to reach all of the allegedly infringing devices, plaintiff adopted an overly broad posture. In doing so, however, plaintiff left itself open to the failure of proof which the Court has found to exist under a proper interpretation of the patent.
In view of this state of the record and the clear and convincing evidence supporting the Court's determination of patent invalidity, the Court has chosen to abstain from making full, detailed findings on the several infringement issues.
The Court finds and concludes that plaintiff has failed to sustain its burden of proving the claim of patent infringement by defendant; and that upon the facts and the law, plaintiff has shown no right to relief as against defendant.
The Court finds and concludes that United States Letters Patent No. 3,003,436 on a "Method and Apparatus for Protecting Vehicle Loads," issued in the name of William H. Peterson on October 10, 1961, is invalid by reason of the obviousness of its subject matter under 35 U.S.C. § 103.
This opinion contains and constitutes the findings of fact and conclusions of law required under Rule 52(a) of the Federal Rules of Civil Procedure.
Judgment will be entered dismissing the complaint with prejudice on the merits.
Although the Court believes that the foregoing opinion contains the findings and conclusions required under Rule 52(a) of the Federal Rules of Civil Procedure and that they adequately reflect the Court's answers to "the factual inquiries necessary to resolve" all the issues, including "the issue of obviousness under § 103," see Kerr v. State Farm Life Insurance Company, et al., 373 F.2d 62 (2d Cir. 1967), any party may submit on notice supplemental or additional findings or conclusions consistent with the foregoing opinion, within ten days after the filing date of this opinion.
Settle judgment on notice within ten days after the filing date of this opinion.