HICKS, Circuit Judge.
Peale-Davis Company (herein called Peale-Davis) sued West Kentucky Coal Company for infringement of certain patents. American Coal Cleaning Corporation (herein called American), the manufacturer of the alleged infringing machines, intervened in the defense.
The claims before us are 14 and 16 of patent No. 1,782,391, November 18, 1930; 35, 43, 45, and 50 of No. 1,786,739, December 30, 1930; 11, 17, and 21 of No. 1,787,340, December 30, 1930; 17 and 18 of No. 1,817,298, August 4, 1931; and 26, 30, and 35 of No. 1,822,840, September 8, 1931. Each patent was issued to Kenneth Davis and assigned to Peale-Davis.
Claims 14 of No. 1,782,391 and 26, 30, and 35 of No. 1,822,840, were for apparatus ; the remaining claims, for a process.
The principal defenses were anticipation, prior use and knowledge, noninvention, and noninfringement.
The court held the process claims void for lack of invention, and the apparatus claims uninfringed.
The case involves a process and mechanisms for “dry cleaning” unsized coal. The process frees it of a certain percentage of slate and other refuse with which it has become mingled during the mining operations.
It was the well-established practice, after coal was brought from the miné, to [656]*656pass it over screens And thus grade it into various sizes. The larger sizes, “lump” and “egg,” were then transferred to cars or bins by conveyors and men stationed alongside removed the slate and other foreign matter by hand. But smaller sizes, ranging from an inch downwardly to slack or dust, could not be cleaned economically by hand, and were cleaned, if at all, by being subjected to some sort of nonmanual separation, as by wet washing jigs, centrifugal separators, or pneumatic tables; but even these treatments did not clean the coal thoroughly, and as large consumers began to buy coal by specifications, calling for smaller and smaller percentages of residual ash, other ways had to be found* of eliminating the refuse from the small-sized coal.
The patentee claims that he has discovered a process and mechanism for cleaning coal as it comes from the mine which eliminates the slow and costly prescreening and sizing, and does a better and Cheaper job. The process is relatively easy to grasp, though the machinery for achieving the end is complicated.
Patentee utilized the fact that various substances of the same relative volume have different weights (i. e., have different specific gravities). As we have intimated, this principle had long been observed in the separation of smaller sizes of coal and slate after screening. Projected into the air or jostled about, materials of different specific gravities tend to stratify, the lighter moving to the top and the heavier gravitating to the bottom: But in thus separating the lighter coal from the heavier slate and other impurities, by means of the pneumatic table, it had been supposed that it was necessary to prescreen the mass so that the treatment could be applied to particles of the same relative size, say one inch to one-half inch, one-half inch to one-fourth inch, one-fourth inch to one-eighth inch, and so on down within the ratio of two to one. It was also supposed that the smaller the classification, that is, the nearer the mass came to slack or dust, the less successful the segregation.
Pneumatic and mechanical separation of coal from its impurities either before or after prescreening involved something of the same equipment. Usually there was a table, however shaped or positioned, with a pervious bottom through which compressed air was forced. The uncleaned coal was delivered to the table at one end through a hopper and while the air was passing up through the floor of the table and the layer of coal, the table was vibrated by an eccentric mechanism (perhaps several hundred times a minute). Sometimes the floor of the table had ribs or ridges (called “riffles” in this art) which aided in vibrating the coal and caught some of the impurities and channeled them to the place of discharge. Thus the bed of coal on. the table was never quiescent, being jiggled by the vibrations and loosened by the air currents. While it was thus blown up and stirred about, it was also activated by gravity, and as the mass settled back each time, the heavier particles, slate and other extraneous matter, tended to settle lower and collect behind the ridges, whence they were propelled in a different direction and discharged in a different place from the lighter chunks and particles of coal which were tumbling off the table.
In patentee’s process the coal was delivered to the table just as it came from the mine and thus the expensive and tedious prescreening which also tended to break up, or degrade, the coal into smaller and less valuable sizes was obviated. The patentee utilized the “fines” (defined as any particles smaller than the two to one ratio of the largest size treated) in the unscreened mass; the theory being that these “fines” confined the air, thus preventing the dissipation which would take place through the interstices of large sized coal, creating a tendency to raise the entire layer on the table, so that the gravitational and vibrational effects necessary to the segregation of the heavier elements could get in their work.
Patent No. 1,786,739, for a process. Claim 43 1 is typical. In this patent the [657]*657bed used was a plane surface dropping away from the hopper. The riffles on it were crosswise of the bed and the vibration was lengthwise and across the riffles. Due to the slope of the bed, the coal, being flotant over the heavier refuse, tended to rise to the top of the layer and to “float” toward the lower end of the table discharging into a chute. The heavier material tended to sink to the bed whence it was propelled by the interaction of the reciprocating motion of the table with the riffles upwardly along the bed surface to the chutes in a direction opposite to the flow of the coal.
The air was regulated so that the strongest lifting currents were near the hopper outlet. The larger pieces of rock settled to the bed and the larger pieces of coal rose to the top, being supported on the heavier materials and blanket of air. The force of the currents was diminished by zones in the direction of the lower end of the bed, so that the mildest air current was near the coal discharge point; thus the finest particles of rock tended to settle to the bed in the zone of least air pressure, leaving the stratum of pure coal superimposed.
Successive inventions introduced refinements in the mechanisms by which the process was practiced.
Patent No. 1,787,340, for a process. In No. 1,786,739, the movements of the coal and refuse were directly opposed to each other; in No. 1,787,340, the courses were athwart each other, but at an angle; the “heavier material” moving “transversely beneath the superposed lighter material.”
The preferred embodiment of the table had a deck higher in the middle and declining slightly to either side,' — -the reciprocation was along the long axis. The riffles were placed on the inclined portion and extended from the hopper for about half the length of the table parallel to, the long axis of the deck and to the direction of reciprocation; and then converged inwardly toward the center or axis of the unit. The heavier materials collected against the riffles but were inched to the far end of the table, away from the hopper, for discharge. The coal spilled down the slopes of the table and over either side. Any coal that rode the full length of the table toward the refuse discharge was deflected to the edges by vertically disposed skimmers, inclined outwardly and forwardly from the central part of the unit toward the edges. These deflectors preferably rested upon the top edges of the riffle plates. The air pressure was strongest at the center of the table and declined through two zones to each edge and was controlled for each zone by different sized meshes in the air pervious bottom.
Claim 21 of the patent is identical with claim 43 of No. 1,786,739, down to the clauses dealing with air regulation. We print the remainder of the claim.2
Patent No. 1,817,298. This was for a mechanism and process for the immediate removal, at several points, of the various sizes of refuse as they were separated out so that the remaining coal was cleaner. In the specifications the patentee said, “I am able on this one table to practically completely separate the coal, rock and bony and other impurities from three inches in dimension down to the finest dust.”
In this patent the deck was flat, with a V-shaped coal discharge groove cut into it from the end away from the hopper, converging at a point about two-thirds of the way to the hopper. The riffles diverged from the center, making an angle with each other slightly greater than the angle of the V we have just described. Thus the refuse was carried to the edges of the table, whence it was discharged through gates on either side occupying about two-thirds of the length of the table. The coal was discharged through the central V, the first coal tumbling into it very soon after leaving the hoppe.r, and after passing over the area of greatest air pressure. The other coal was subjected to [658]*658further cleaning as it passed along the two arms of the deck over the areas of diminishing air pressure.
Claim 17 3 is typical.
Patent No. 1,822,840. The table is deeply cleft almost four-fifths of its length. In the preferred form the riffles are parallel with the outside edges. The coal is discharged over the two outer edges and the refuse into the V. Its deck is level and the V is cut so deeply it practically divides the deck; in fact, it is spoken of as a “duplex table.” It was preferably upwardly and forwardly inclined and the reciprocation was along its long axis. The V is provided with a vertical retaining wall along its edges to prevent the discharge of the superior stratum of coal into the refuse recess. Along the V, orifices are provided in this wall at deck level for the discharge of heavy material which has settled to the deck. The riffles stop short of the wall so that there is passage along its side for the concentration and conduction of material through the orifices. Toward the end of the table the V diverges sharply and a battery of discharge orifices are provided in the wall. The air pressure is heaviest under the hopper and along the V, grading off into five other zones toward the spillage edges. The air chest goes under the entire table and is split into compartments by lengthwise partitions, portions of which reach deck level and other portions stop short thereof to permit passage of air over the top into other compartments. The air in these compartments is directed upwardly to the table by horizontally mounted vanes operated either by handles or through' doors in the floor.
The claims in suit are for a mechanism and No. 26 is printed.4
Patent No. 1,782,391. It antedates the others and deals with a mechanism and method for air control on cleaning tables, which do not relate especially to the unsized processes of cleaning coal. The single plane table was sloped from the hopper both to the front and to the side. The riffles were disposed longitudinally and preferably paralleled. The table was provided with, a plurality of perforations disposed in the space between the riffles. The specifications provided,' — “A layer or layers of fabric or wire mesh, or other material may be employed * * * to keep the material from dropping into the perforations. * * * ”
The air chamber underneath the deck was divided into a plurality of compartments by partitions extending beneath the riffles. At the air supply end, each of the compartments was supplied with a damper or valve pivotally mounted on the floor of the chamber for regulating the air supply. The pivotal pin projected up in each case and was provided with a handle for turn[659]*659ing the valve ¿nd indicating its position. Each of these compartments communicated through a perforated bulk head with a larger chamber connected by a flexible pipe with the bellows, fan, or other source of air supply.
A second set of dampers, of any number and position, with handles projecting downwardly were supplied within the compartments to reduce or cut off the supply of air’ to the parts of the chamber beyond them.
With reference to the action of the air on the materials to be cleaned, the specifications say: “ * * * The separation of the intermixed materials is effected by the combined flotation of the lighter materials by the air pressure in connection with a vibratory motion of the table or support acting to throw the lighter particles across the riffles from one depression to another, and the heavier particles along between the riffles.”
Claim 14 is for the mechanism and claim 16 for the process. Each is quoted below:
“14. A mechanism for separating intermixed, divided materials including in combination an air-pervious table, means for maintaining thereon a bed of materials extending substantially continuously there-over, means for progressively stratifying the' materials in superposed strata according to their specific gravities and for progressing the settled heavier material and a continuous, flbtant superior stratum of lighter material to separate places of discharge including devices for vibrating the table, separating partitions solely on the surface of the table for directing the settled heavier material transversely of the direction of flow of the flotant lighter material, and mechanism for passing lifting air currents upwardly through the table and the bed of materials thereon, said air supply mechanism comprising a main air chamber under steady air pressure connected to the table and means for substantially independently controlling the supply of air from the main chamber to different areal subdivisions of the table located at different places both longitudinally and laterally of the table.”
“16. The process of separating intermixed divided materials which comprises maintaining a bed of the materials in substantially continuous progression along an air-pervious table, subjecting the bed to mechanical vibration and upwardly-directed, lifting air currents to stratify the materials, progressing settled heavier material and superposed lighter material to different places of discharge, maintaining a steady supply of air under pressure for the air-pervious table, and regulating the admission of air from said supply to relatively small areas of the table located at different places longitudinally and laterally thereof.”
Invention: Claim 14 of No. 1,782,-391.
The uniqueness of this claim lies in the disposal of air chambers and dampers to effect a gradation in the air currents. We have been cited to nothing that closely resembles it in the prior art, although all sorts of devices for controlling the air were disclosed. The claim represents a step forward and discloses a more convenient and flexible method of control than the punched metal baffle plates, perforated decks, etc., and we think with the District Court that it should stand, but should be limited to a narrow range of equivalents.
Claims 26, 30 and 35 of No. 1,822,840. We have more difficulty with these apparatus claims. Davis was familiar with the Y-table described in the Sutton patent No. 1,710,521, April 23, 1929, before he applied for this ’ patent. The Davis tables are modified in many respects from the Y, but if the arms are removed from the Y-tables a resemblance is apparent. The coal discharges to the outside as in the Y-table, the riffles incline inwardly in one form and the refuse goes to the center. The table has been flattened but it has the same slight upward slope from the feed end and there _ is the same relative disposal of air pressures over the table as in the Y-table.
On the other hand, Davis has added the deeply penetrating V-shaped groove in the center of his table to catch the refuse, the walls of which, spreading outward, tend to shove the coal to the spillage edges. This is important. The Sutton patent carried the refuse along the open passageways between the riffles for the length of the table. In Davis the separation between coal and refuse starts much sooner. The coal begins to tumble and be shifted over almost at once. The refuse works to the center and begins to discharge halfway down the V. This is important in a mechanism designed to ac[660]*660complish a separation in one operation and we think Davis made a patentable advance upon the Sutton Y-table.
The Process Claims: We think that Davis impressed the coal mining art with the boldness of his conception in dumping run of mine coal onto his table and adjusting the forces actuating the table to clean it and in utilizing the “fines” as a step in the process whereby slow and costly prescreening was obviated.
If we had any doubt that the Davis method claims were patentable it would be resolved by the widespread and sustained success of the Peale-Davis plants, comprising eleven installations since the latter part of 1927, independent of their own— plants treating almost every conceivable range of coal, from five inches to zero, in Lincoln Gas Plant, to one-half inch to zero at the Ashland Plant. All but one of these plants have been very successful commercially and represent installation costs of $50,000 to $250,000 each.
Prior Art: 'Some of the devices are of passing importance only. Hooper, letters patent No. 189,734, issued April 17, 1877, reveals the age of some of the elements of the Davis patents — the pervious bed, air chamber, bellows, diagonal channels for the collection of separated ore (forerunner of riffles), skimmer bars, etc., all, as early as. 1877, used in a dry ore separator. In Card, letters patent No. 448,472, issued March 17, 1891, we find a dry ore concentrator to which to the ore bed and bellows of Hooper had been added a “beater” or vibratory agitator. These two inventions are sufficient to show the lineage of the art.
Emery Patent No. 1,272,642, July 16, 1918. The specifications state that the invention was designed, “ * * * to effect the separation of the various constituents of ore * * * as * * * galena from quartz. * * * The invention is also applicable to the separation of the various constituents of other granular mixtures than ores, such, for instance, as slate from coal, etc.”
' The specifications also say that the machine achieves a classification and concentration of the unsized products in one operation; that is, it requires “no preliminary screen sizing (except between very wide limits). * * * ”
Appellees claim that this invention, particularly, discloses the Davis method, but a casual inspection of this patent indicates that though .it uses the familiar air currents, vibration, and gravitational roll, the actual segregation is effected in an area designedly kept free from the upwardly directed air currents. In the preferred form this is “dead space” and comprises more than half of the area of the table. In an alternative mechanism the “dead space” is made pervious, but to down currents. The process does not pass lifting currents throughout substantially all parts of the bed, nor does it regulate the air action and continue it throughout the travel of the bed. It is not the Davis process and only remotely resembles it.
Tyler patent No. 1,262,603, April 9, 1918. As an anticipation, we think this patent fails as completely as Emery. The air was directed upwardly, not through the pervious deck, but through relatively widely scattered upraised nozzles, which permitted large dead air pockets in between, which were essential to its functioning.
The Sutton, Steele & Steele patents, called herein the Sutton patents. Appellees rely heavily upon some of the later ones of this series. In our view of the Davis process we find it unnecessary to consider in detail the earlier ones, for in No.- 1,699,382, January 15, 1929, upon an application filed December 17, 1925, it was said in the specifications: “While our improvements are capable of being used for the separation of masses of different kinds of materials they are particularly adapted for the separation of the particles of the masses of coal, the said masses of coal ranging through the smaller sizes, for instance from one-sixteenth of an inch in size to one-one-hundredths of an inch. It is well known that all coal treated on the table shown in our said co-pending application” (now Patent No. 1,710,521 filed on January 19, 1922), “and treated on our type of table shown in‘ various prior patents, is first screened into a number of sizes, and each size run over the table that is provided with the proper adjustments for that particular size of coal. The reason for this is that should the coal without being first sized be passed over our tables at one time separation would naturally be very indifferent since the table is working on the principle of differences in specific gravity in the particles that form the mass. However with the best screening methods this condition exists to some extent, being due largely to the attrition of the particles. The present improvements [661]*661will compensate however for the ordinary amount of under-sized particles that appear in the mass being separated. The Y-type of table” (No. 1,710,521) “has been working on a product of coal far in excess of these conditions, the products having the wide range in size hereinbefore referred to.” (Italics ours.)
In an earlier Sutton patent, No. 1,073,-644, September 23, 1913, there is a reference to the cleaning of particles “varying in size from one-sixth to one-eighteenth of an inch,” depending on specific gravity, but the Sutton patents do not disclose a method or process for cleaning run of mine or unscreened coal.
Infringement: After Peale-Davis began the installation of its plants, American began using an en masse process. O’Toole, President of American, describes the process in the Transaction of the American Institute of Mining and Metallurgical Engineers — Coal Division 1931. He said: “We have done considerable experimenting in cleaning coal dry on our separators. We have progressed from ten tons per hour when we first started to almost every quantity up to one hundred twenty-five to one hundred fifty tons per hour, depending upon the size and kind of coal treated. We have changed our practice from pre-screening to mass treatment. * *
Appellees claim “that the Davis patents in suit are for a one table process' * * * while defendants’ process * * requires a plurality of tables and screens; * * * and that after the first partical (sic) separation” defendants’ process “involves certain steps that are not even contemplated by the Davis patents in suit. * * * »»
Although American seemed customarily to have used three separators or tables, and West Kentucky Coal Company used three tables, their claim is weakened by a statement in their “American Bulletin 32” for 1932, that the number of separators used is governed partly by the results desired and partly by the character and distribution of the impurities. The quotation said: “ * * * Size ranges from three inches to zero are being cleaned successfully by treatment on a single American Pneumatic Separator without screening.”
It remains for us to describe somewhat more particularly appellees’ mechanism and process
In answer to interrogatory 16, they say: “Each of defendant’s pneumatic separators has a deck 14 feet 6 inches long and 6 feet wide. It is longitudinally reciprocable and has side spillage edges substantially parallel to the line of reciprocation. The sides of the deck slope, downwardly from the middle constituting in effect two decks. The coal flows by gravity in opposite directions from the middle upon these two.deck sideá. Each side of the deck has a plurality of riffles extending longitudinally and inwardly toward the middle. The middle portion of both sides óf the deck is longitudinally unobstructed until the forward part of the deck is reached. At the forward part of the deck there are two banking bars or walls, each extending from the middle of the deck forwardly and laterally to one of the side edges.”
This is appellees’ so-called “R” table, a refinement of the Sutton Y. It has similarities to the mechanism described in Davis No. 1,822,840, except that the Davis table is flat and has a more deeply notched V-shaped center. The interrogatories bring out another less apparent difference, affecting the regulation of the air under the table. There were numerous louvres, regulable from the outside while the table was in motion. It is asserted in the answer to interrogatory 35-F (and the mechanism is discernible in Appellees’ Exhibits Nos. 67 and 68) “the air pressures cannot be effected (sic) by setting said baffles in their extreme open and extreme closed positions due to the fact that there is an open space between the baffle plates” (louvres) “and the pervious deck. * * ” A difference is claimed here, since the air did not pass through the louvres into a closed compartment, as in claim 14 of Davis No. 1,782,391, but rather into a large chamber into which air jets were fed from numerous louvres.
There are other differences which we do not regard it necessary to detail here.
Infringement of Apparatus Claims: The apparatus claims of Davis No. 1,822,-840, are very generally worded. We have previously noticed certain resemblances between the Davis table and the Sutton Y-table. We think the claims should have no blanket coverage but should be limited to the disclosures of the specifications and drawings. It is clear that the table herein claimed was level and depended either upon skimmer bars or the inertia of the [662]*662coal, as the table was reciprocated under it, to deliver the coal to the edges. In the alleged infringing R model appellees resorted to the sloping deck of Sutton and the force of gravity to bring the coal to the edges. We concur in the holding of the court that none of these apparatus claims were infringed.
We think Claim 14 of Davis No. 1,-782,391 from which the R-table differed even more .decidedly is likewise uninfringed.
Infringement of Process Claims: The nature of appellees’ process is simple. There were usually three R-tables connected by conveyor belts, the first effecting a primary separation of the coarser coal and coarser débris. The débris was discarded. Between these extremes of pure coal and pure débris was a product spoken of as “middlings” composed partly of coal and partly of foreign matter. This material and the clean coal, as they came off the table, were passed over a screen which took out the larger coal. Some of this was conveyed to a separate bin and some to an “assembly bin” to be mingled with clean coal of other sizes as it came from the later separators and'screens. The middlings which passed through the first screen were conveyed to a second separating table where pure coal and pure débris and further middlings were segregated. The pure débris was led off, the pure coal and middlings, as before, were passed over a screen. The middlings -from this screen were conveyed to a third separating table and treated as before.
Appellees’ process was revealed in “American Bulletin 31,” 1931, by a diagram which showed crude coal three inches to zero in diameter being passed on to the first separator. The refuse therefrom was led off to the refuse bin and all other products went to a one and one-half inch screen, whence clean coal three inches to one and one-half inches went into a bin of “stove” size or onto a conveyor to the “assembled sizes” bin. The one and one-half inch to zero crude coal passed over the second separator, refuse therefrom to bin and other products over three-fourths of an inch screen. Part of clean coal therefrom, one and one-half inch to three-fourths inch in size, went to “nut” bin and remainder to “assembled sizes.” The three-fourths inch to zero crude coal from the second screen was conveyed to the third separator, thence refuse to discard and clean coal three-fourths inch to zero, partially to “slack” bin and partially to “assembled sizes.” This diagram was labeled “Typical Flow Diagram American Pneumatic Separation Plant.” It was purely schematic; there were other diagrams in evidence disclosing the treatment of other sizes and there was testimony that sometimes crude coal up to six inches was treated, and there is the quotation from Bulletin 32 that with some coal successful cleaning was achieved “by treatment on a single American Pneumatic Separator without screening.”
This statement is particularly damaging to appellees’ contention that theirs “is not a one table process.”
The importance of appellant’s process lies in the fact that, in one continuous operation, it cleans a bed of raw coal as it comes from the mine; or coal that has not been closely screened; and the essence of it lies in the utilization of “fines” to build up air pressure, and thereby produce the necessary disturbance in the layer on the table.
Concomitant with the use of the “fines” was the co-ordinated activity of the riffles, the reciprocation, the shading of the air action, and the force of gravity. We think that appellees’ process was substantially identical with it and that it infringes claims 35, 43, 45, and 50 of Davis patent No. 1,786,729. Appellees’ use of three tables does not disclose a different method. To the contrary, it does disclose in the final analysis the practice of appellant’s method upon each of the three. The screening in between the treatments adds nothing, other than contributing to the simplification of the problem of adjusting the air pressure, etc., on the second and third tables due to the presence of more uniform sizes in the coal to be cleaned. Appellees cannot avoid infringement by máking an unimportant addition to the practice of appellant’s process.
Claim 11 of No. 1,787,340, is somewhat indefinite as to size range of intermixed materials covered by it but we think appellees’ process comes within this claim even though it is thus indefinite and we rule it infringed to avoid hair-splitting as to size ranges covered by the process.
We think claims 17 and 21 of this patent are also infringed.
Claim 17 of patent No. 1,817,298 covered the process of cleaning an intermix[663]*663ture of “relatively fine particles and large pieces such as are present in mine coal, which has not been subjected to close preliminary size classification.” This claim as well as No. 18 of the same patent we rule to be infringed by appellees’ process.
Claim 16 of No. 1,782,391 was for a process of separating “intermixed materials” not “fines.” The forces used were all disclosed in Sutton and if there was any infringement it was of the improvement in areal regulation of the admission of air to the table. This was done after a fashion in Sutton. Since there was no claim here for the whole process dealing with “fines” and since in the Sutton patent areal regulation of a sort was disclosed, and since the method practiced by appellees was less precise, we think this claim should be narrowly construed and is hence not infringed.
The decree is reversed, with directions to enter a decree in accordance with this opinion.