Opinion
No. 4645.
November 8, 1927.
Appeal from the District Court of the United States for the Eastern District of Michigan; Arthur J. Tuttle, Judge.
Suit in equity by the Haynes Stellite Company against Percy C. Chesterfield and another. Decree for defendants, and complainant appeals. Reversed.
For opinion below, see 8 F.2d 765.
Chas. Neave, of New York City (Maxwell Barus, of New York City, on the brief), for appellant.
Ridsdale Ellis and Chas. W. Hills, both of Chicago, Ill. (Chas. W. Hills, Jr., of Chicago, Ill., and Myron J. Dikeman, of Detroit, Mich., on the brief), for appellees.
Before DENISON and MOORMAN, Circuit Judges, and GORE, District Judge.
This is the usual infringement suit, based upon patent No. 1,057,423, issued April 1, 1913, to Ellwood Haynes for a metal alloy. The patented article in its commercial form is called stellite, and the suit was brought by the manufacturing company, which owns the patent, against the makers of a more or less similar alloy, called by the trade-name "Chesterfield." The District Court thought that the patent sued upon was invalid as to the broader claims, and as to the narrower ones, if valid, was not infringed. Both parties find the utility of their alloys in the field of machine tools, which are used upon lathes or similar machines for cutting and shaping iron or steel parts. There is no great difficulty in making a suitable cutting tool of ordinary steel, so long as it is so used that, because of slow speed or for other reasons, it does not develop much heat; but when such a tool runs at higher speed, and gets hot, it becomes soft and useless. To meet this difficulty the art, some 25 or 30 years ago, developed what was called high-speed steel. This could be made so hard by a heat treatment that it would maintain its cutting edge fairly well up to a temperature of about 600 degrees. This enabled tools made of it to be run at about twice the speed possible with ordinary steel; but, when about this temperature was reached, the efficiency rapidly fell off and soon ceased. It was then necessary that the tool be retreated and rehardened before it could be used again.
Nothing better for this purpose than this high-speed steel was in use or was practically known when Haynes devised the patented alloy. It was a ternary compound, composed (in the specific form now involved) of the easily fusible metal cobalt and the refractory metals chromium and tungsten. The proportions were to be varied as directed in the specifications, but for use as machine tools it was recommended that the tungsten be not less than 25 per cent. nor more than 50 per cent. of the whole, that the chromium be 15 per cent., and the remainder be cobalt. This alloy, in this recommended form, was put upon the market by Haynes and his company, and, either in the original or in a slightly modified form, it and similar compositions have superseded everything else for many classes of metal cutting operations. In suitable applications, the use of this alloy permits double the speed possible with the best high-speed steel, and its utility, translated into money savings from speed of operations in all branches of metal manufacture, is rightly characterized by counsel as stupendous. Its novel characteristic, as compared with its only initial competitor — high-speed steel — is called "red hardness," meaning that, although red hot, it will retain its cutting capacity, while the steel in the same situation becomes "red soft." This quality of red hardness, from the degree of about 600 up to about 1000 (centigrade) was entirely novel in any commercial existing alloys, and the stated ternary compound was unknown to the art, and so was itself novel, unless by the effect of the items to be mentioned. We do not overlook that existing compositions had some cutting capacity, even at a low red heat; but they did not have a high, or even valuable, efficiency over much, if any, of the above-stated temperature range.
Haynes' specification shows that the invention in its broadest aspect consisted in uniting cobalt with any two of the metals of the chromium group. This group was said to comprise chromium, tungsten, molybdenum, and uranium. The specification gives particular attention to chromium and tungsten, and claim 8 calls for "a metal alloy composed of cobalt, chromium, and tungsten in the proportion of from 5 per cent. to 60 per cent. of tungsten, the remainder of the alloy being cobalt and chromium." This claim is infringed, if any one is. We see no practical occasion for considering the validity of claims which might be satisfied by other metals of the chromium group, although, since tungsten stands in the specification as the equivalent of other metals of the group, we are required to observe the supposed anticipations which refer to the group, regardless of specific mention of tungsten or chromium. We turn to these supposed anticipating or limiting instances.
A former patent of Haynes is much relied upon. By his patent No. 1,299,404 he covered his invention in what has come to be known as stainless steel, and by uniting with iron a relatively small quantity of chromium and a trifle of carbon he produced an alloy which was ductile enough to be forged into table knives and similar articles, but was highly resistant to the corrosion which affects common iron and steel. This had great commercial success, and the patent was sustained in American Co. v. Ludlum Steel Co. (C.C.A.) 290 F. 103. In an earlier effort to make an alloy of the same class, which would be ductile enough for the same uses, but which yet should take a high polish and be immune to stain, Haynes developed two patents, one of which called for a composition of nickel and chromium, or some other metal of the chromium group, and the other of which described a composition of cobalt and chromium or some other metal of the chromium group. These two patents (873,745 and 873,746) disclosed that, in Haynes' judgment, for producing an alloy having both the required stain and rust resisting qualities, cobalt and nickel were more or less equivalent as a first member of the composition, and that any one of the four chromium group metals was the equivalent of any other one of that group as the second member of the composition. Both patents contemplate only a binary compound, and there is no suggestion that two metals of the chromium group could be used so as to make a ternary compound. It is likely enough that Haynes then supposed that he might use in this composition one or more of these chromium metals, and that whether he used one or two would not make any difference; but the patent calls for only one of them. He did not, at that time, use more than one; and when later he made the ternary compound he found that he had a new material which, because of its hardness, would have been useless for some of the purposes of his binary patents, and which developed a new characteristic before unknown to the art. We are unable to give Haynes' binary patents any substantial force as against the patentability or breadth of his ternary patent.
The Leffler British patent, 16,829 of 1891, disclosed a plan for hardening steel sufficiently for use as a cutting tool and without any heat treatment. He used pig iron as a base, into which he says he introduced two or more of any of a long list of metals — nickel, cobalt, chromium, tungsten, molybdenum, uranium, etc. This seems to be one of those general descriptions, which are intended to be all-embracing and which produce nothing concrete. Haynes' idea is not present. Leffler thought he could use two or more, and seemingly any two or more, of his list of metals. He would therefore have satisfied his disclosure if he had put nickel and cobalt into his iron without any chromium, or if he had put in tungsten and chromium without any cobalt, or if he had put in copper and lead without cobalt or chromium. Such hit or miss reference does not disclose the combination of cobalt and two of the chromium group. Further, these other metals were not to be so used by Leffler as to make an alloy composed of them. They were instead rather trifling ingredients, modifying the characteristics of the iron with which they were united. The specific disclosures of the three compositions which he suggested contemplate from 87 to 96 per cent. of iron, from 1 to 6 per cent. of tungsten, 2 per cent. of nickel, and from .5 to .3 per cent. of chromium. One of his specific combinations contains also .5 of molybdenum, no one of them has any cobalt, and while cobalt and nickel were broadly analogous in the binary compounds, they are not completely equivalent as to the patented ternary compound. We think it not permissible to consider this Leffler patent as fairly disclosing an alloy composed of "cobalt with two or more metals of the chromium group."
Thompson's British patent, No. 7,847 of 1895, is also urged as having a limiting effect upon the patent in suit. We do not think it has. It does not disclose or definitely lead to the red hardness which is the outstanding characteristic of the Haynes ternary patent. Thompson, like Leffler, contemplates only putting into iron or steel some 2 to 3 per cent. of his compound by way of alloy. His first result, like his final one, is by no means a metal alloy composed of cobalt and two refractory metals. Thompson again shows a vague proposition by one who names a list of metals, among which he is to have unrestricted choice, and who points out none of the distinctions which must govern that choice in order to approximate Haynes' alloy.
Marsh's British patent (Boult), No. 2,129 of 1906, superficially suggests complete anticipation, because in his first general explanation the patentee says that his discovery is "that any one or more of the metals of what is termed the chromium group, when mixed with nickel or cobalt, produces an alloy possessing properties, etc."; but further examination demonstrates that Marsh did not disclose the Haynes invention. Marsh was devising an alloy suitable to be drawn into wire or similar form to constitute a resistance element in electrical heating apparatus. It must therefore be quite easily workable, and a far less degree of hardness than is the minimum characteristic of Haynes' ternary compound in the form preferred for a lathe tool would have been fatal to Marsh's purpose. Only in his introductory paragraph does he suggest using "any one or more" of the chromium group; in his detailed specification he points out that he combines his cobalt or nickel with "any one" of the chromium group in order to get his desired result. In his detailed statement he was correct, at least as to certain ones of the group. In his vague and general statement, he stated only his supposition, and the supposition was erroneous; or, as counsel truly say, this was not only a mere prophecy, but it was a false prophecy. The first reference to "one or more," instead of "any one," was apparently merely casual and inadvertent. To this patent, as well as to the other British patents mentioned, the language of Judge Shipman, in Westinghouse Co. v. Great Northern Co. (C.C.A.) 88 F. 258, 263, is most appropriate: "The prophetical suggestions in English patents of what can be done, when no one has ever tested by actual and hard experience and under the stress of competition the truth of these suggestions, or the practical difficulties in the way of their accomplishment, or even whether the suggestions are feasible, do not carry conviction of the truth of these frequent and vague statements. * * * The result [otherwise] reached is not shaken by merely a single sentence in the English patent."
We have considered the several other objections urged against the validity of the patent, but regard them, upon the whole, as no more forceful than those already discussed. It is true that claim 8, by contemplating a tungsten content as low as 5 per cent., includes grades of the alloy which would not have the characteristic red hardness and which would be workable by forging or otherwise. Claim 8, as drawn, might not be sustainable if the compound, in the softer grades, was old; but, while the red hardness gives the sharpest novelty and the extreme commercial value, we must conclude that the alloy itself, to the extent of the whole scope of this claim, was novel, and even in the softer grades was sufficiently useful to support patentability.
The only remaining question in the case depends upon the fact that defendant, in its alloy, uses some carbon and some nickel in addition to cobalt, chromium, and tungsten. Infringement is therefore denied. The question must be whether these ingredients constitute merely additions to the patented alloy, preserving the characteristics thereof, and adding some effects that may be improvements, or whether the use of the five, instead of the three, elements makes a new and different combination. Defendant's claim is that the carbon is essential to make an effective cutting tool, and that plaintiff has recognized this by including carbon in the composition of its commercial product. Plaintiff's reply is that its alloy, as first put upon the market, was made precisely like the specification and contained no carbon; that this was a successful article, and, as then constructed and as samples have now been reconstructed, its cutting tools have the highest efficiency and have the full capacity of red hardness; that, when the alloy came to be made in larger quantities, it was found advisable to add an exceedingly small amount of carbon for the purpose of reducing the fusing temperature of the other ingredients and thus cheapening the cost of manufacture; and that the small amount of carbon so added has no material effect (unless in some conditions, as a mere improvement) upon the performance or qualities of the otherwise complete patented alloy, within the field where the alloy has its greatest commercial value.
It is enough to say that we are fairly well satisfied, from the expert evidence and the practical tests made, that plaintiff is right, and that the carbon is a mere addition (even if sometimes an improvement) to the patented combination, the employment of which by defendant does not avoid infringement. The same conclusion must be reached as to the use of the fraction of nickel which the defendant substitutes for some of the cobalt (14 per cent. of the whole, one-third of the cobalt-nickel total). Nickel and cobalt are not, precisely speaking, equivalents in making up the patented combination, but they have some of the same qualities; and the fraction of nickel substituted by defendant we are satisfied does not substantially change the character of the cobalt-chromium-tungsten compound.
The decree below should be reversed, and the usual interlocutory decree for injunction and accounting be entered, based upon claim 8. The decree may be silent as to the other claims. We see no sufficient object in pursuing such distinctions, if any, as there may be.