delivered the opinion of the court:
This is an appeal from the decision of the Board of Appeals of the United States Patent Oflice affirming the decision of the examiner rejecting certain claims of appellant’s application for patent relating to a method of welding steel, particularly martensitic steel. Six claims, numbered 1 to 6, inclusive, were involved before the tribunals of the Patent Office and were included in the appeal to us, but appellant withdrew claims 1, 3, and 4 from the appeal, so that we have to consider only claims 2, 5, and 6.
*872Claim 2, which is the broadest of the three claims, reads as follows:
2. A method of welding- martensitic steel, including heating said steel around .the welding zone to temperatures sufficiently high so that when said zone is heated during welding to temperatures forming austenite it cools upon completion of welding at a rate effecting direct decomposition of austenite to a product other than martensite, and welding said steel at said zone while said steel around said zone is heated to' the first named temperatures, said steel being treated prior to its welding to have a determined tensile strength and the first named temperature being- insufficiently high to materially alter said strength.
Claims 5 and 6 specify a range of temperatures from 350 to 1,100 degrees Fahrenheit to which the parts of the metal adjacent the weld are heated. Claim 6, in addition, defines the weld resulting from the process as having “a tensile strength exceeding 185,000 lbs. per sq. in.”
The brief of the Solicitor for the Patent Office summarizes the teaching of appellant’s specification as follows:
The application relates to a method of welding heat-treated martensitic steel in such manner that the welded joint is at least as strong as the welded members. Martensitic steel is a carbon steel which is formed by rapid quenching. The steel thus formed is somewhat brittle but may be heat-treated1-by tempering so that it has a tensile strength of 135,000 to 250,000 lbs. xaer square inch. The members to be welded are made of this -tempered martensitic steel. If the members were welded in the ordinary manner by heating merely the abutting edges to' a molten condition, the weld would cool very rapidly because the heat would be conducted away from the weld by the adjacent cold portions of the members. By this rapid cooling the material forming the weld would be transformed to ordinary martensitic steel, which, as stated before, is somewhat brittle. The weld would therefore not be as strong as the tempered martensitic steel members. Appellant’s object was to cure this defect by preventing the formation of the brittle martensitic steel in the weld as it cools and causing instead the formation of a stronger steel in the weld. As pointed out above, martensitic steel is formed by rapid quenching of heated steel. If the steel is cooled more slowly by keeping the temperature in the neighborhood of 1,000°-1,200° F. for a short time, approximately 10 seconds, the steel is transformed to a variety known as pearllte. This, according to appellant, is at least as strong as tempered martensitic steel. In his method therefore appellant retards, the cooling of the weld material so that it is transformed to pearlite instead of martensite. To retard the rate of cooling, appellant heats not only the abutting edges of the members to be welded but also to a somewhat lower temperature the adjacent regions of the members. The heat present in these adjacent regions prevents the undesired rapid cooling of the weld.
The following references were cited in the board’s decision:
Anderson, 1,930,847, October 17, 1933.
Bain et al., 1,924,099, August 29-, 1933.
Anderson, 1,972,509, September 4, 1934.
Metals Handbook, 1936 Edition. Published by American Society for Metals, Cleveland, Ohio. (Copy in Div. 8.) Pages 109-110.
The patent to Bain et al. does not relate to the welding art but to the thermal hardening of steel. It is described in part in appellant’s brief before us as follows:
*873This patent is concerned with the hardening of martensitic steel by heating it to an austenitic condition, quenching it 'at a rate at least equal its critical cooling rate, which would mean that if it were quenched to a low enough temperature martensite would be produced, and then, prior to reaching such a lower temperature, holding the temperature of the steel within a range of from 300° to slightly .less than 1,000° F., the result being the production of a new structure having material advantages over tempered martensite and which has since become known as bainite. The patent shows the now well-known S-curve of austenite transformation * * *, but it does not graphically show the cooling rates superimposed on this curve. However, reference to the specification of the patent shows that it is predicated upon clearing what is called the nose of the S-curve, which is in the vicinity of 1,100° F. in the case of the curve of the patent, so that the steel can enter the slow austenite transforming range which the patent features.
The patent to Anderson, No. 1,930,847, discloses a method and apparatus for welding, and embraces both method and apparatus claims. The Anderson patent, No. 1,972,509, is entitled “Method of Welding Seams Along Curving Surfaces” and is limited to the method. It recites, substantially, the method steps of his earlier patent (to the application for which reference is made in the application for the later patent) and some additional steps.
The Anderson patents are particularly concerned with the butt-welding of opposed end edges of pipe, or tubular sections of metal. Neither of them specifically names iron or steel, but appellant does not seem to question the fact that the process they describe is applicable to steel. He insists, in effect, however, that the steel of the articles contemplated by Anderson was low carbon steel and did not present the problem which appellant solved. '
As best we can determine from the record before us, the examiner rejected the claims on the Anderson patents in view of Bain et al., citing the Metal Handbook article as showing that the idea of preheating a high carbon steel (this in answer to appellant’s contention respecting the steel of Anderson being low carbon steel) prior to welding was old in the art.
The board’s discussion of the merits of the case is quite brief. It merely said:
As stated by tbe examiner, the metallurgical problem presented here is deemed to be fully set forth in the patent to Bain et al. This patent is concerned" with martensitic steel. The patents to Anderson disclose the necessity for preheating the metal before the welding operation so as to insure a weld of substantially homogeneous composition and in which the crystalline character of the metal does not vary abruptly. These patents are concerned with the welding of pipe, and while no specific reference is made to iron or steel, still it is very probable that the metal contemplated was a mild steel of low carbon content of the type usually employed in making so-called iron pipe. By reason of the low darbon content, the problem presented in these patents was not the same as in the present ease, but from the description in patent No. 1,972,509 it is apparent that the patentee was attempting to achieve analogous results.
*874It is clear from the Anderson patents that his process involved, first, the preheating of sections of pipe lying adjacent to the end edges to be welded, the preheating extending back some distance from the edges; and second, the heating of the seam formed by the junction of the edges to welding temperature, the method being continuous. The Anderson process is fairly described in the brief of the Solicitor for the Patent Office as follows:
As will be evident from Fig. 2 of tbe patent [No. 1,930,847], tbe parts to be welded are passed beneath a series of beating jets 22 arranged to converge toward tbe seam. Tbe welding takes place where the jets converge on the seam. Tbe beat of tbe weld, however, is prevented from being conducted rapidly away by the beat already present in the adjacent regions, which beat was previously applied by tbe diverging series of jets 24.
The result obtained is stated in the patent as follows:
* * * while beat continues to be conducted to tbe relatively cool regions away from which tbe application of beat is being displaced, still, due to the beat previously applied at these portions, walls of hot metal are created which render impossible any large difference in temperature between fusing metal and nearby solid metal. This effectively reduces conduction away from the highly heated area of fusing metal, thereby promoting penetration, and eliminates the usual demarkation between parent metal and the metal of the completed weld. On the contrary, because of the advantageous temperature gradient which is maintained, both the internal structure across the weld into the parent metal, as well as the external formation are much improved over results formerly obtained.
So, it is obvious from the prior art that there was nothing new in the step of heating the parts of metal to be welded for a considerable distance on each side of the weld in order to retard leakage of heat from the welded section. That, in the final analysis, appears to us to be the essence of appellant’s process.
It is true that in appellant’s case the process is applied to a type of metal different from that to which Anderson’s process is applied, but, so far as the claims on appeal are concerned, the alleged invention lies in the process and not in the resultant product.
We do not overlook the fact that there is no teaching in the Anderson patents of the range of temperatures applied, while in appellant’s claims 5 and 6 the temperatures are defined as being in the range of from 850 to 1,100 degrees Fahrenheit, but this range is shown in the Bain et al. process for the thermal hardening of steel, which (while not describing a welding process) describes, as stated by the examiner, “the metallurgy of the effect of heating a martensitic steel.” This effect is fairly stated in the excerpt from appellant’s brief above quoted.
We- fail to see wherein it was error on the part of the board to affirm, as by implication it clearly did, the holding of the examiner that “the Bain patent in the metallurgical art would teach the welding art which temperatures to use when welding martensitic steel in order to obtain after cooling a weld of the desirable character.”
*875Having the process of Anderson in mind, it would seem that if a martensitic steel is to be welded, one skilled in the art would readily apply the teaching of Bain et al. to the problem.
So, notwithstanding appellant’s arguments respecting various details, such as the terms “fusion” and- “annealing,” we are not convinced that his process involved invention.
The appeal as to claims 1, 3, and 4 is dismissed, and the decision of the Board of Appeals as to claims 2, 5, and 6 is affirmed.