m SCSI Wm mm m ■b^HhIIH W" BB!HWWWJIMW « S wa m yn f Vt m* w W > wrruw ' w HrtfWtflti yiwiiKjfflPi'UYr rr nrKr-fJuinrtffl JBfcag '•••f •■• nzm # ;; Class __„ Book L COPYRIGHT DEPOSIT Digitized by the Internet Archive in 2011 with funding from The Library of Congress http://www.archive.org/details/highspeedsteeldeOObeck HIGH-SPEED STEEL Published by the McGraw-Hill BookCompany New 'York. (Successors to tke Book. Departments of tke McGraw Publishing Company Hill Publishing" Company Publishers of Books for Electrical World The Engineering - and Mining Journal The Engineering Record rbwer and The Engineer Electric Railway Journal American Machinist Characteristic Colors (in full daylight) Pale Yellow Stra Dark Straw Brown Yellow Light Purple Purple-blue Full Blue Polish Blue Dark Blue DEGREES Bright Red in the Dark Red in Twilight Nascent Red — * Red- Dark Red Cherry Bright Cherry Dull Orange - Light Orange - Lemon Light Straw White Brilliant White Dazzling White Characteristic Phenomena — ^ Softening Begins Range of Ar > (Softening Critical I Points) in Carbon Steel Softeuing Completed ^Softening Begins j Range of Ar •s in High Speed Steel Softening Completed >> [Range of Arj.Ar.,, Xand Ar, in Carbon ISteel Nascent Cherry — »- 5 - === 800 IT [Range of Ar,,Ar, 1500 = — N and Ar, inHigh- ^| [speed Steel (heating) > Hardening Range High-speed Steel Frontispiece Heating Phenomena, Fahrenheit and Centigrade Scales. HIGH-SPEED STEEL THE DEVELOPMENT, NATURE, TREATMENT, AND USE OF HIGH-SPEED STEELS, TOGETHER WITH SOME SUGGESTIONS AS TO THE PROBLEMS INVOLVED IN THEIR USE BY O. M. BECKER INDUSTRIAL ENGINEER McGRAW-HILL BOOK COMPANY 239 WEST 39th STREET, NEW YORK 6 BOUVERIE ST., LONDON, E. C. 1910 < $> ■I & ^p° Copyright, 1910, BY THE McGRAW-HILL BOOK COMPANY Stanhope iptcsa F. H. GILSON COMPANY BOSTON, U.S.A. ©CU2689H7 INTRODUCTION It is rather unusual that any considerable time should elapse between the announcement of a great discovery and the appearance of a book deal- ing specifically and comprehensively with the subject of that discovery; and especially is this true when the idea or invention has such a far-reaching, even revolutionary, influence in its particular department of the world's activities as is the case with high-speed steels. Nevertheless during the decade and more since Fred W. Taylor and Maunsel White unexpectedly stumbled upon the high-heat treatment for tungsten and related steels, and developed their high-speed steel, nothing has been published which could be called at all adequate as a treatment of the subject, except pos- sibly Mr. Taylor's address before the American Society of Mechanical En- gineers on the occasion of his inauguration as president, at the December, 1906, meeting; and the articles hereafter mentioned. This address or report is indeed a monumental work, including as it does the results of his experiences and researches, and those of his co-laborers, in the carrying on of his investigations. These researches, it would seem, have been more extensive and thorough than any others yet attempted, covering the gen- eral subject of the new steels. Other investigations, covering limited portions of the field, like those of Dr. H. C. H. Carpenter, say, have been exhaustive as far as they extended, and also are very valuable for the light they throw upon the nature and the possibilities of high-speed steel. The Taylor report confines itself largely to the experiences of its author and his associates, and unfortunately, has not a good index. There has been a pretty general feeling for some time that a work should be available which would cover comprehensively the whole subject of high-speed steels, presenting in an understandable way a general view of it in such form as to be easily accessible for reference. This was in the mind of the present writer when he was invited some five years ago to contribute to the Engineering Magazine a series of papers dealing with the subject. The articles were written and printed, 1 and, as far as appears, formed the most extensive account of the new steels, as adaptable to use in productive industry, which had been published up to that time. 1 Engineering Magazine, New York and London. Issues of September, October, November, and December, 1905 ; May, June, and August, 1906. iv INTRODUCTION Since the original articles - appeared much new light has been thrown upon certain aspects of the subject. Mr. Taylor's address has been pub- lished, and many sporadic contributions of greater or less value have ap- peared in the technical periodicals. Furthermore, important developments have been worked out concerning which little or nothing at all has been heretofore printed. While therefore this book is in a sense based upon the series of articles already mentioned, it is by no means the same. Such of the original material as has proved fundamental has been retained and brought into accord with present knowledge and practice. Much has been added touching the historical and theoretical aspects of high-speed and other tool steels. The purpose has been to include all that might be of interest in connection with the subject and to present an accurate con- spectus of the present state of knowledge concerning the new steels. The entire subject is so recent, however, and astonishing developments have followed one another with such bewildering rapidity, that the accom- plishing of this purpose presented many difficulties. Almost every steel man knows the practical impossibility of obtaining detailed information concern- ing tool steels which is absolutely accurate and reliable. The personal equation and a multitude of other factors commonly accounted negligible, so much affect results that the assertion might be safely hazarded that all conclusions as to method and means in steel treatment are subject to allowances for these elements. Thus, for illustration, two different plants operated by the same concern make large use of the barium process in hardening. At one the temperature commonly maintained and' indicated by the pyrometer is 1200 degrees C. (2200 F.), while at the other an instrument of the same make and calibrated with equal care uniformly indicates only 1070 degrees C. (1950 F.); and yet as far as can be seen by the eye of an experienced operator, familiar with both plants, the two baths are kept at identical temperatures. Certain it is that results equally good are obtained at both plants. Similar difficulties arise in connection with methods and apparatus. Thus, it is maintained with much fervor by some that the only furnace giving good results is coke-fired; while others insist with equal fervor that oil or gas fired furnaces alone can be depended upon. Similarly, con- clusions as to superiority of one or another brand of steel over others are almost uniformly based on insufficient grounds or arrived at under condi- tions not precisely duplicated (or not likely to be duplicated) elsewhere. Great care has been taken to insure absolute accuracy of statement, and definiteness in every respect. The reader must remember, however, that in view of the considerations just set forth, there may be some disagreement with methods here proposed and conclusions presented. The endeavor has been not only to cover, in the chapters dealing with the practical han- dling of the new steels, all points likely to come up, and to indicate clearly just what is believed to be best practice at the present time; but, where INTRODUCTION V there is diversity of opinion, to describe all practical methods shown to yield satisfactory results. In conclusion the author wishes to express his appreciation of the assist- ance rendered by Mr. Walter Brown and Mr. D. G. Clark. Except for their encouragement and patient assistance during the early days of the work the task, if undertaken at all, might never have been finished. Grate- ful acknowledgment is also made of the interest manifested and assistance rendered in the ways of suggestions, criticism, permission to use material, and proof reading by Mr. Fred W. Taylor, Dr. H. C. H. Carpenter, Mr. George H. Paltridge, Dr. Bradley Stoughton, and others. Dr. Carpenter was especially helpful with suggestions in connection with the chapter on the Nature and Characteristics of the New Steels; and Dr. Stoughton was kind enough to revise portions of the same. Acknowledgment for the use of illustrations is made in the appropriate places. O. M. BECKER. Chicago, III., July 1, 1910. CONTENTS Page Introduction Section I. The Development and Nature of High-Speed Steels. Early Tool Steels 1 Self-Hardening and High-Speed Steels . . . 13 Nature and Characteristics of the New Steels 22 Still Newer Steels 49 Section II. Making the Steels and Tools. Process of Making High-Speed Steel 55 Forging the Tools 64 Hardening — The High Treatment Practically Applied 78 Hardening — The Barium Chloride Process 105 Tempering 121 Annealing 129 Grinding 134 Ascertaining and Regulating Temperatures 155 Miscellaneous Observations on the Making of High-Speed Tools 178 Section III. The Tool at Work. Range of Utility of High-Speed Steel 186 Conditions of Maximum Effect 212 Speeds and Feeds, and Related Matters 235 Section IV. The New Requirements. Fundamental Considerations in the Design of the New Tools 252 The New Machine Requirements 275 Remodeling an Old Equipment 295 Section V. Problems Involved in the Use of High-Speed Steels. Statement of the Problem 300 Making a Beginning 319 Section VI. Appendices. A. Analyses of High-Speed and Special Steels 330 B. Taylor on Hardening High-Speed Tools 332 C. Reference Table for Determining Cutting Speeds 334 D. Diagram for Determination of Time of Turning Operations 336 E. Summary of Metal Removed and Cost of Removal 338 F. Taylor Tables of Practicable Speeds 339 Index 345 HIGH-SPEED STEEL. CHAPTER I. THE DEVELOPMENT AND NATURE OF HIGH-SPEED STEELS. EARLY TOOL STEELS. The Earliest Tools. — That the extraction of iron from its ores was practiced long before historical times, is evidenced by many remains of prehistoric forges and smelters. Not a few implements of that material whose antiquity cannot be doubted, also have been found. Scriptural authority ascribes the first forging of iron to Tubal Cain, whose time "is generally fixed at about 38 centuries B. C. Whether or not steel dates back so far as that, it is quite certain that steel imple- ments must have been used not much later, in the construction of certain of the very earliest ancient monuments known to us; for tools of soft iron could not have cut and carved the very hard rocks of which they. were constructed. In connection with the building of the pyramids Fig. 1. The monumental carvings of Egypt are conclusive evidence that the ancient Egyptians possessed good steel tools, which alone could have cut the hard stones smoothly. The pictures themselves sometimes, as in this case, indicate the kind of tools in use and the methods of using them. (date fixed at about three thousand B.C.) Herodotus speaks of the immense sums of money which. must have been spent for the " iron " with which the builders worked; and in the Great Pyramid there was found a fragment of an iron tool which must be at least five thousand years old, containing a small percentage of carbon. There is good reason for believing also that the working of steel as well as of iron was known to and practiced by the early Greeks and others, perhaps as far back as ten centuries B. C. Ancient Steel Making. — As to the method making of the common steels of antiquity, little is certainly known. The iron to be converted into steel appears to have been made in a sort of crude Catalan forge; and as iron has a strong affinity for carbon, it is probable that in forg- ing iron which had been long in contact with the fuel and therefore 1 2 HIGH-SPEED STEEL had taken up some carbon, it would be found somewhat harder and stiffer than common iron, and on that account more suitable for tools. Among the very ancient Egyptians, it is said, pieces of meteoric iron were heated in close contact with the fuel and kept at a high tempera- ture, just below the melting point, for a long time. The absorption of carbon was of course sufficient to make a superior iron, or steel, of the piece so treated. The next step in the development of steel no doubt was the accidental dropping into water of a red-hot tool of some kind. The superior hardness and usefulness of such a tool would certainly lead to further experiment and more careful treatment. Whether or not this was the actual course of events in the development of ordinary steel, it is certain the ancients possessed some remarkably good steel implements and knew pretty well how to use them. The ancient method of working metals was by forging, chiefly. If this did not give a suitable result, or if a sharp cutting edge was required, recourse was had to abrasives. Machinery being practically unknown, it was impossible for primitive men to do much in the way of metal cutting, although indeed it seems that graving or hand cutting of some kind was practised almost if not quite as far back as the time of Tubal Cain. Certain it is that the art was practiced among the Chinese long before the Christian era began. Ancient Metal Cutting. — Not however until the beginning of what has aptly been called the age of machinery, which dates' back scarcely Permission of The Iron and Steel Magazine. Fig. 2. A primitive furnace for smelting iron and producing steel. The ore and fuel lie in the fore- ground. In charging the furnace, these were carried up the rude steps carved in the hillside against which the furnace was erected. more than a century, did the real development of metal cutting begin. A crude form of the lathe had been known for a good many centuries, as long ago indeed as the days of Solon, in the sixth century B. C; but THE DEVELOPMENT OF HIGH-SPEED STEELS 3 this improved but little upon the tedious and laborious hand-cutting process. The arrival of steam power made possible the development of relatively heavy and powerful machinery, and permitted the large development of cutting tools. Wootz, Damascus and Toledo. — From the earliest times the quality of steels varied greatly. The celebrated wootz steels of India, and Damascus made at the Syrian city of the same name, and later at Toledo, in Spain, were examples of superior steels susceptible of extraordinary tempering. They both, however, were crucible steels, and the latter is also said to have contained strong traces of tungsten, nickel, man- ganese, and similar elements important in the new high-speed steels. Production of Wootz. — Wootz is a steel of high quality, the manu- facture of which reaches back to very ancient times. Aristotle, the Fig. 3. Section of a crude Catalan forge, still to be seen in remote places, in which steel is often made directly from the ore. The charcoal and ore, the latter in small lumps, are charged at the level of the floor or a little above. As the reduction proceeds, a more or less steely lump, the "loupe," collects at the bottom. Greek philosopher, describes it as nearly as 350 B. C. or thereabouts, and tells us how it was made. Other ancient writers in the subse- quent centuries likewise make mention of it. This steel still is (or at any rate until recently was) fabricated by some of the remote mountain peoples in the north of India. The method of manufacture is essentially the same as that described by Aristotle nearly twenty- three centuries ago. First the iron is extracted from the ore in a sort of rude clay Catalan forge, perhaps built against the side of a hill. The result is a lump of sponge iron, which is hammered to make it more dense and to rid it of as much of the unmelted ore and cinder as may be. The lump is then broken into smaller pieces, hammered again, 4 HIGH-SPEED STEEL and afterwards placed in a specially prepared small clay crucible, holding about a pound of metal. Along with the iron is put some finely divided wood, and a few green leaves, the latter laid at the top of the charge. The crucibles are luted up and placed in a diminutive furnace of clay, generally a bare two feet in diameter, conical, and sometimes partly below the surface of the ground. The blast is furnished by a pair of skins blowing through a bamboo and clay tuyere. After several hours the crucible is tested for complete fusion of its contents, by being shaken; and if the melting is complete, the crucible is allowed to cool, to be later broken open. The result, if successful, is a small lump of exceedingly fine, smooth-surfaced steel, which takes a remarkable temper. This steel has been used almost exclusively for swords as far back as tradition runs; and it has been said of Wootz blades, as also of Damascus, that when properly tempered and skill- fully handled they would cleave a bar of iron without losing edge. It has been declared also that a blade of wootz has cleanly cut a wisp of silk floss tossed into the air. Crucible Steel among the Chinese. — This product of ancient skill seems to have been little if at all known, as an article of commerce, to those nations who have done most toward the advancement of civilization through the development of the steel and iron-making art. Nor, apparently, was it known to them that steel was produced in eastern Asia contemporaneously with, and as now seems established, long before wootz was made in India. Recent archaeological discoveries indicate that a kind of crucible steel was known to and made by the Chinese probably many centuries before the Indian product. The method by which the Chinese steel was produced is not yet known. It probably was something like that practiced in India; the more so since there are indications that both these peoples may have learned the process from a common source, a people living in the plateau between the two lands. Manufacture and Nature of Damascus. — Damascus steel likewise was a crucible steel. It differed from wootz however, and from other steels as well, in one respect. Those familiar with its appearance are aware of the curious veining or figuring seen on the surface. This characteristic, adding greatly to the beauty of a finished surface, is due to thin strips of soft iron alternating with similar strips of steel, and these thoroughly welded, twisted and otherwise completely wrought together. This steel was first made, as far as known, at the ancient Syrian city (said to be the oldest existing city in the world) whence it derives its name, and doubtless also is of great antiquity. The Scrip- tural reference to " iron from the north," together with other allusions in Scripture and the historical references to the Chalybes (whence is derived the Greek and Latin names for steel) indicate that the region THE DEVELOPMENT OF HIGH-SPEED STEELS O north of Damascus, if not that city itself, was a renowned center for steel manufacture hundreds of years before the beginning of the Chris- tian era. Virtues of Ancient Steels. — The virtues of Damascus steel were made famous throughout the western world by the crusaders, and its manu- facture was at a later period established at Toledo, in Spain. Toledo blades came to have a repute almost as great as those made at Damas- cus. It is stated that they possessed tempering qualities as remarkable as those of wootz, and that they were sometimes packed in boxes curled up like clock springs of our day. Whether or not these ancient steels possessed all the remarkable qualities attributed to them, they certainly were marvels of their time; and indeed they remain marvels in our own time. Nothing superior to them, in respect of tempering quality, has been produced by modern methods of steel making. It is certain that all of the steels mentioned were at some time or other used for tools of various sorts, very likely for use in the metal-cutting arts. For the latter purpose, however, they possessed no striking advantage over other steels. Variability of Primitive Steels. — The quality of ancient steels, and of those produced up to a comparatively recent time, varied greatly. Considering the methods of manufacture and the diversity of ores used, this is not surprising. Some ores would contain elements wanting in others; and in the process of forging, or perhaps of extraction, certain tools would be in closer and longer contact with the fuel and would therefore absorb more carbon. Modern manufacturing methods make it possible to eliminate all these uncertainties, and it is comparatively easy to produce continuously steel of practically constant quality. Development of Metal Cutting. — The development of metal cutting was, until a few years ago, brought about almost wholly through the evolution of the machine by which the tool was made to do its work, and scarcely at all through the development of excellence in the tool itself, except in so far as it was found that varying shapes gave varying results. The steel, its nature, and the method of treatment, remained much the same as for centuries before. With the development of powerful machinery, however, it was soon found that there was a limit to the amount of work to be got from a tool for cutting metal. The tenacity with which the particles of a homogeneous mass of steel, iron, or similar tough metal cohere makes it no slight matter to drive a tool into the mass or force off a portion of it. A small graver in the hand can easily make a scratch in the surface of an iron plate; but to remove a chip say a quarter of an inch by three-quarters, a cut by no means unusual nowadays, involves the consumption of an astonishing amount of power. Now the energy used up in making a cut, is partly, of course/converted into latent energy 6 HIGHSPEED STEEL stored in the chip with its changed form and changed relation of con- stituent particles; and partly into sensible heat at and near the cutting point of the tool, which heat is taken up by the tool, the chip, and the piece being machined. The chip being relatively small and continu- ously changing in respect to the particles of metal at the cutting point, rarely gets so hot as to show a color higher than deep blue, and not often that. The piece machined is relatively large, and readily absorbs and conducts away its portion of the heat, which is a small part only of that generated. The tool, however, is continuously at work, and absorbs a good deal of the heat generated. Now if the tool be worked heavily, which is to say usually if the cutting speed be relatively high, the cutting edge quickly gets so hot as to draw the temper and make the ordinary carbon steel tool useless. Endurance Limit of Tools. — Evidently there is a limit to the amount of work such a tool is capable of doing; and this limit, the snail's pace at which it has heretofore been necessary to carry on metal-cutting operations, has been an anomaly in modern industry the chief character- istics of which are magnitude and speed. There is a vim and vigor about wood-working operations as seen in a modern shop, which is exhilarating. A large spindle, for example, is shaped while revolving at the rate of two or three thousand turns per minute, the rate, of course, depending somewhat upon the size and nature oj: the wood. A steel shaft of similar diameter revolving against an ordinary tool at a very small fraction of the same number of turns would almost instantly " burn " it. Maximum Speed with Carbon Tools. — Carbon steel, as heretofore used in tools, no matter how well hardened, has not enough toughness and hardness to withstand the rubbing of the chip for any considerable length of time, even when not run fast enough to affect the temper. The tool therefore dulls; and this dulling proceeds in a sort of geomet- rical ratio as the cutting speed increases, being augmented by the draw- ing of the temper which accompanies rapid cutting. The speed in all metal-cutting operations has therefore had to be comparatively 'slow, no matter how powerful might be the machines in use. Thirty feet of chip per minute, as any machinist knows, has been considered rather good work; while fifty feet per minute has been very unusual. Under ordinary circumstances the management of a shop was pretty well satisfied if the machine tools could maintain an average speed of twenty to twenty- five feet per minute. Such deliberation, necessary though it has been, is depressing in this era. A creeping mass of metal turning leisurely round and round or moving back and forth, as has been customary in the average shop, is quite out of harmony with the modern spirit of expedition and hurry. But while a few dreamed of the possibilities of cutting metal, some time in the future, with something of the vim with which wood can be THE DEVELOPMENT OF HIGHSPEED STEELS 7 cut; and while machines had been developed so tremendously as to leave scarcely anything to be desired in that respect; nevertheless the ultimate limit seemed to have been reached. But it had not. Similarity of Ancient and Modern Steels. — For at least a thousand years, and probably for several thousand, there had been no single important advance, no one striking development, in the nature and characteristics of steel, in respect to metal-cutting qualities at any rate. The property of becoming hard, possessed in common by all steels, and distinguishing them from ordinary iron, is due to the presence of carbon diffused throughout the mass of the metal. How the presence of the particles of carbon brings about the virtue of hardening is yet a matter of discussion. The Modern Science of Steel Making. — Modern steel making is a fact only because the science of chemistry, itself scarcely a century and a half old, has made it possible to understand that there is an affinity of certain elements for certain others and that under given conditions exactly the same combinations can be expected in chemical compounds and alloys. , The prehistoric steel makers had no idea that in firing iron with certain fuels they were carbonizing it, actually forming of the iron and fuel a new substance which contained besides iron the same element which in one form constitutes charcoal, in another graphite, and in a third, diamond. That, however, is exactly what they did. Blister and Double Steel. — The method of the very ancient steel makers, except as already noted, was essentially the same as that com- monly in use up to comparatively modern times. The bar to be car- bonized was heated in close contact with charcoal or other suitable fuel. In later times this was done in a sort of double muffle furnace, and the cementation or carbonization was more even and complete. Steels thus made however, were found to be only skin deep, so to speak; for since trie carbon merely soaked in and combined with the iron nearest to it, evidently the central portion was less completely carbonized than the exterior, and except possibly in very thin bars was generally quite devoid of the hardening element, even after ten or twelve days' treat- ment. In later times, when the muffle furnace came into use, steel thus made was known as blister steel, from the blisters or scales appear- ing on its surface during the process of carbonization. This steel is not dense and uniform enough for fine tools. It was known for many centuries, however, that hammering (and later rolling) so as to " work " it thoroughly, greatly improved the quality and usefulness for tool making of such " cementation " steels. An improvement upon this method was that of breaking the bars into lengths, bundling them, and then welding together. The " shear steel " thus produced was of much better texture and uniformity, but was not so good as the " double " or " double shear " steel which was made by repeating the process a 8 HIGHSPEED STEEL second time. Damascus seems to have been made somewhat in this fashion, but after the desired uniformity had been obtained in the steel itself, thin layers of the steel and of fine iron seem to have been again welded together. The grainy or " watered " appearance of Damascus is said to be due to this streaking with iron. Revival of the Crucible Method. — " Shear " or " double " steel was of course a great deal better adapted to edge tools than any produced by the earlier methods; but it still had the disadvantage of being more or less streaked and spotted, for no amount of hammering or rolling could entirely eliminate the inequalities of carbonization by the cemen- tation process. It was not until about the middle of the eighteenth century that a method of overcoming this defect was discovered. About that time one, Huntsman, was astonishing other makers of blister steel by the absolutely uniform texture of his steel. It lacked the " seams " or streaks characteristic of the other steels of that day, and hardened uniformly all the way through. Keeping a process or method secret was at that time considered the only way of distancing competitors, and Huntsman and his workmen managed to keep their trade secret to themselves until the envy of a competitor, so the story runs, imposed upon their humanity and learned the secret. On pretense of seeking shelter one stormy night this competitor, according to the tradition, appeared at the forge where the wonderful steel was made, and was at last admitted for humanity's sake. What his expectant but astonished eyes beheld was so absurdly simple that he may well have wondered why he and others had not thought of it also. It was in fact nothing but the melting of the broken pieces of blister steel in a crucible. Of course the steel made in such a way would be uniform, for each crucible, at any rate. Had the spy but known it, however, he beheld a process which in its essential features was centuries old; for as the reader has already seen, crucible steel has been known in some parts of the world since time immemorial. Later Methods. — Steel made by Huntsman's method came to be known as " crucible " or " cast " steel. The first name is still considered a sort of trade-mark for steel made in this way, though there are many other steels nowadays made in pots or crucibles and in one sense there- fore entitled to be so called. This crucible steel at once came into favor and held its position as the tool steel par excellence until the introduc- tion of mushet steel and the subsequent development of the Taylor- White process. The result of these advances was something essentially different and in every way superior for most, if not indeed for all, tools used in the metal-cutting arts, and perhaps also for all cutting pur- poses. The only important change in manufacturing crucible steel was made by the elder Mushet about the beginning of the last century. Instead of melting blister steel in the crucible, he used refined iron THE DEVELOPMENT OF HIGH-SPEED STEELS 9 (scrap or bar) mixed with some carbonaceous compound. The soft iron was thus carbonized in about the same way as the sponge iron in making wootz. Steel produced in this -way, however, had not the ex- cellence of that made in India; nor even of the blister-crucible steel of Huntsman. Some fairly good steel is thus made, but usually it needs to be thoroughly " worked " to make it dense and good enough for fine tools. Later Mushet mixed pig-iron with the contents of the crucible; and this is still usually clone except when the blister-crucible steel is required. A peculiarity of crucible steel is that it must be " dead-melted " or else it is liable to be more or less porous and otherwise imperfect. This consists merely in allowing the melted steel to remain fluid in the crucible for a half hour or more, before pouring. Open Hearth Steel. — Many attempts have been made to produce as good a steel by other methods, whereby the cementation process could Fig. 4. Longitudinal section through a Siemens regenerative open hearth furnace. To start the fur- nace a wood fire is burned in the two chambers at one end of the furnace (G and A). When these are red hot, a current of air is passed through the brick checkerwork in A, and up a flue to M; a current of fuel gas is passed through the checkerwork in G, and up a flue to TV. The gas and air, thus preheated, meet at K and fill the furnace chamber with a hot flame, which radiates its heat to the metal on the hearth H. The hot gases then pass down to chambers A and G on the opposite end of the furnace, and store their waste heat in the brick checkerwork there. At intervals of 15 to 20 minutes, the currents of air and gas are reversed in direction, and enter the furnace through the alternate pair of regenerative chambers, taking from them the stored-up heat to create a still more intense flame over the hearth. be avoided. There have been, and still are, many forms of open hearth furnaces which produce a fair quality of steel direct from pig or from refined iron; and the bessemer process has been greatly improved and modified in so far as results are concerned. But neither of these type processes produces a satisfactory tool steel, though some very good results have been secured from certain carefully made open-hearth steels. 10 HIGH-SPEED STEEL In the open hearth furnace (Fig. 4) from five to nearly one hun- dred tons of steel can be made at a heat, whereas the crucible process makes but fifty to one hundred pounds per pot. Essentially the open hearth process consists in remelting old steel scrap and mixing with it pig iron, which is either melted at the same time or else brought in the still liquid condition from the blast furnace in which it is made. The proportion of pig iron will vary all the way from ten to nearly one hun- dred per cent, depending on the state of the market, the quality of steel to be made, and other like conditions. Pig iron is an impure iron con- taining three to four per cent of carbon, one to two per cent or more of silicon, sometimes phosphorus, manganese and other impurities. The impurities are diluted by the mixture of steel scrap, and are further reduced through oxidation by iron ores added to the furnace charge for the purpose. When the bath is purified to the desired point, man- ganese and a little silicon are added to rid it of dissolved oxygen. Open hearth steel is used in immense quantities in machinery and structural work, but makes tools inferior to those of crucible steel. 1 The heats are large, which necessitates pouring them into large-sized ingots, and these are softer and inferior in the center after cooling. The man- ganese added to cure the superoxidation is not a perfect antidote, and it is impossible to prevent imperfections in steel made in this way, or to obtain such a uniformity of hardness as is required for the best tools. Bessemer Steel. — In the Bessemer process, liquid pig iron is brought from the blast furnace and poured into the converter (see Fig. 5), The Fig. 5. Section of a bessemer converter. Silicon, manganese and carbon are removed from the molten pig iron by blowing air through it. The converter is mounted on trunnions, T, one of which is so made that a blast of air passes through it and along a duct, D, to the bottom, whence it comes up through openings, A, in the bottom. When the metal is judged to be in proper condition, the converter is rotated on the trunnions and the charge emptied. 1 That is, inferior to crucible steel made in the manner described, for steels have been put upon the market under that name which are quite inferior. Though possibly they have been melted in a crucible, the name is misleading, and usually is intended to be so. THE DEVELOPMENT OF HIGH-SPEED STEELS 11 carbon and silicon in the metal are then Burned out by blowing a spray of cold air bubbles through the molten bath. This combustion supplies the heat which not only keeps the metal liquid, but even raises its tem- perature some hundreds of degrees, and white hot flame pours from the Fig. 6. The bessemer converter in action. mouth of the converter during the ten minutes or so while the purifi- cation proceeds. After practically all of the carbon and silicon are removed, the predetermined dose of melted iron rich in carbon, man- ganese and silicon (known as the " recarburizer ") is poured into the converter to rid the metal of superoxidation and give it the desired grade of carbon, and it is then cast into ingots. Bessemer steel is still more liable to irregularities and imperfections than open hearth, and is unsuitable for good grades of tool steels. ft HIGH-SPEED STEEL Electric Steel. — The production of steel in the electric furnace is the only other important process in use today, and by this means electrical energy is substituted for fuel in producing the necessary heat. The metal is entirely protected from oxidation in this process and is capable of a purification unattainable by any of the others. The steel is said to be superior even to crucible steel, and to be made at a less cost. CHAPTER II. SELF-HARDENING 1 AND HIGH-SPEED STEELS. Manganese-Bessemer. — The addition of manganese to the contents of the bessemer converter made it possible to work the steel freely while hot and helped give bessemer steel its well known properties. It was Robert F. Mushet, himself interested in steel manufacture, who sug- gested the addition of manganese in the making of bessemer. He did not stop with the results his suggestion quickly brought about, but continued making experiments for the improvement of ordinary steels. He had at this time no idea of improving steels for use in tools, particularly; but was working primarily to get the best possible steel for ordinary use. While carrying on these experiments, however, he made a discovery of far-reaching importance to all those industries in which metal cutting is practiced. He discovered self-hardening steel. Discovery of Mushet Steel. — Ordinary steel, owing its hardening property to the presence of carbon, has been hardened from time im- memorial by quenching in water while at a red heat, as is well known. If allowed to cool slowly, as in the air, it is too soft for use in tools. During the course of his experiments, sometime in 1868, Mushet found that one of his bars seemed to have the property of becoming hard after heating, without the usual quenching. This circumstance was not merely singular; it was astonishing, and contrary to all previous experience. Possibly it marked a new epoch in steel making. Analysis of the bar behaving so singularly showed that it contained a percentage of tungsten. Properties of Tungsten Steels. — Not only did the bar containing tungsten harden without the customary quenching, but it was actually harder than ordinary steel which had been quenched. It occurred to Mushet that this extraordinary circumstance might be turned to advan- tage in the production of superior tool steel, and he accordingly set him- self to developing tungsten steel with this end in view. The result of experimenting with hundreds of metal mixtures in the crucible was a 1 Mushet steels soon came to be known in England as "self-hardening." After a time the term "air-hardening" was used more or less in the United States along with "self-hardening." In this book the term "self-hardening" will be used to refer to mushet steel, though of course high-speed steels also are partially self-hard. On the continent high-speed steels are frequently known as "rapid steels." 13 14 HIGH-SPEED STEEL steel alloy much more satisfactory than any other then in use, which possessed the property of becoming very hard by mere exposure to the air. Improvements in Air-Quenching Steels. — It was not until after the steel had come into somewhat general use that it was discovered by Mr. Henry Gladwin, then associated with Mr. Mushet at the Clyde Steel Works, Sheffield; and by several other engineers almost at the same time, that still better results could be obtained if the cutting portion were reheated and then cooled in an air blast. This discovery, in Mr. Gladwin's case, at any rate, was the result of laying some bars of mushet steel on the earth floor of a smithy, near the door. A draft swept over the cooling bars, and they were later found to be superior to bars cooled where there was no draft. Later experiments showed that cooling in an air blast was still better; and that further improve- ment in the quality of tools could be made by bringing the color to a full scaling or almost yellow heat during the re-heating. This, however, was not usually done by toolsmiths, and in consequence most users failed to work mushet tools at their highest efficiency. Mushet Steel in Engineering. — The new steel was immediately put upon the market under the name " R. Mushet's Special Steel." The company organized for its manufacture, and sale, however, did not succeed well in business; and some three years later the production of mushet steel was taken over by Samuel Osborn & Co., Ltd., at the Clyde Works, Sheffield. The wide introduction of the new steel into engineering works, and its imitation under the name of air- or self- hardening steel quickly followed. A substantial advance had been made in the art of cutting metals. It was possible to turn and plane (at first the use of mushet steel was limited to these operations) at double or triple the former speeds; and to machine pieces formerly quite too hard for the tools available or so hard as to make the cost of operation prohibitive. Even after their general use in engineering works, mushet tools were but little used for increasing speeds — most usually only to save frequent grindings or to permit doing jobs previously impossible. It was not until a full quarter century after mushet or self-hardening steel had become an established fact in engineering that the marvelous, and in the light of all previous experience paradoxical, properties latent in it were clearly appreciated, and the industrial world caught a glimpse of what promised to be a revolution in machine shop methods. The Taylor- White Investigations. — The discovery of the possibilities in tungsten steel, like that of the nature of the steel itself, was for- tuitous, if indeed not accidental. In both cases the circumstances that led to the discovery were quite undesigned, and the discovery merely incidental to something else. As far back as 1894 Mr. Fred W. SELF-HARDENING AND HIGH-SPEED STEELS 15 Taylor began experimenting with mushet and other self-hardening steels with a view to determining which was best suited to special kinds of work. This was but a single feature of his program of improving shop efficiency and of determining a logical system of shop management. Shortly after taking charge of the Bethlehem works in 1898 he associated with himself Mr. Maunsel White and others, for better prosecuting the work in hand. After careful tests had been carried on for a time it was decided that a certain make of steel, with proper heating in the tempering process, could be run at higher speed than any of the others; and it was thereupon decided to adopt this make for exclusive use in the shops. , In order to demonstrate their superior efficiency to all the foremen and thus be sure of hearty co-operation in making a change of so great magnitude as was involved in changing a large proportion of the tools then in use, a number of tools of various kinds of steel were ordered carefully dressed, tempered, and ground to exactly the same shape. The foremen were then assembled to witness the comparative perform- ances of the tools. To the astonishment, and we may well believe chagrin, of the demonstrators, the tools made of the selected steel failed to make a good showing. In fact they proved to be inferior to any others in the lot. That is, they could not be worked at as high a speed. Very naturally so unexpected a circumstance would arouse the curi- osity and interest of such keen investigators as were Taylor and White. It had to be accounted for; and so another investigation was set on foot. The range of heating, in the case of carbon steel tools, as is well under- stood, is rather narrow. Air-hardening steels have a still narrower but higher range; and the excellence of a tool depends upon the care exercised in heating it to just the necessary temperature when " draw- ing " or tempering. The first thought that occurred to Taylor and his assistants naturally was that the heat treatment of the tools which failed had been faulty; that very likely they had been under-heated. Whether this was so, or not, seems not to have been definitely deter- mined. One thing however, was determined; namely, that a series of experiments should be undertaken to find out just what would be the effects of various degrees of heating, ranging all the way from a black to temperatures considerably beyond what had been previously thought permissible. The High Heat Treatment. — The results of the experiments were star- tling indeed. When the investigators were in the midst of the work laid out, it was realized that they had made a discovery which upset all previous beliefs as to the effects of heat upon steel, and which was apparently bound to bring about ultimately a revolution in machine shop practice. It was nothing less, indeed, than that steels of the tungsten class instead of being ruined by high heats, were actually 16 HIGH-SPEED STEEL improved so greatly that cutting speeds became possible which pre- viously had been only dreamed of; the discovery, in fact, of the high- speed qualities inherent in these alloy steels when subjected to the super heat treatment. Nothing surprising was noticed until tools were heated considerably higher than had been customary, so high indeed that in the light of all experience the treatment was ruinous. Tools thus superheated were not ruined, but on the contrary stood up to their work better than those given the usual treatment; and apparently the higher the heat treat- ment, the better the tool. For unnumbered centuries it had been believed that steel must not be heated beyond a red; but here were tools which not only were not ruined, but which got better the higher they were heated. The heating, it was found, could actually be carried up to the melting point; and a tool so treated would cut more metal and do it more rapidly than one not raised to so high a temperature. The deterioration of tools which had been heated to near 875 degrees C. (1600 F.) was not surprising; for it was previously well understood CD a, to 00 g s 800 825 850 875 900 925 950 975 1000 1025 1050 Degiees Cent. Fig. 7. Curve showing the influence of the high heat treatment on tungsten steels. that all steels were damaged by being heated so high as this — though indeed, Mushet recommended heating his air-hardening steel to a full scaling or almost yellow heat, which is not far from 1200 degrees C. (2200 F.). This deterioration is shown in Figure 7, which indicates the relative cutting speed possible with mushet steel when heated to tem- peratures varying from 800 to 1050 degrees C. (1450 to 1900 F.). The surprise came when the tools were heated to beyond 925 degrees C. (1700 F.), for with each very slight increase of temperature used in hardening, the cutting power was increased to an extraordinary extent. Improving Air-Hardening Steel. — Tools of this sort were not, however, wholly satisfactory. The cut was rough, and it was by no means cer- tain that the alloy used was the best for this sort of treatment. The SELF-HARDENING AND HIGH-SPEED STEELS 17 discoverers, not satisfied with the results already attained, began to inves- tigate the effect of varying the proportions of the alloy and of the intro- duction of other elements. The outcome of a large number of trial mixtures was a steel capable of doing from three to six times as much work as had previously been possible, and which required for the develop- ment of its greatest efficiency a heat treatment which would utterly ruin ordinary steels. True, it has been pointed out that tungsten and other hardening elements besides carbon were present in some steels of very ancient manufacture, and the tungsten-chromium-manganese steels with which these experiments were carried on had been known for some years. The method of high-heat treatment nevertheless was certainly new, though it would seem that the Mushet experiments and recommended practice, if they did not quite anticipate the Taylor discoveries, ap- proached very close to them. Even if it had been known, the combina- tions of alloys in the ancient steels referred to were quite incapable of developing the powers which modern high-speed steels acquire through the high-heat process, and the old mushet steels needed modifying and improving to adapt them in .the highest degree to the new method of treatment. Rivalry in Experimentation. — For some time the discoverers managed to keep their process to themselves and to the shops to which they sold the rights; but a discovery of such far-reaching importance was bound to become known. Indeed they had no intention of keeping it secret except as a means whereby they might be enabled to carry on further investigations. As soon as the nature of the new steel and the processes of treating them became generally known to the technical and engineer- ing world, and the possibilities of high-speed cutting were in some degree apprehended, manufacturers of tool steel on both continents at once began to vie with one another in their efforts still further to perfect the tungsten steels and to enlarge the range of their usefulness. In the aggregate an immense amount of money has been spent in experimenta- tion with a great variety of mixtures and methods of manufacture. One manufacturer alone tried over two hundred different mixtures and another worked some four years before the product was thought satis- factory. Others have no doubt carried on experiments on a scale equally large, and possibly larger. At the time of this writing (1909) there are perhaps a hundred different brands of high-speed steel upon the market, and new brands of the so-called "new," " improved," or " superior " high-speed steels are multiplying at a rate which bids fair to double the number within a short time. Faults of Early High-Speed Steels. — The first high-speed steels, though of astonishing cutting and wearing qualities, were not adapted to finishing and other fine work. Their coarse, granular structure, 18 HIGHSPEED STEEL perhaps, did not take a cutting edge such as would leave a good finish; and so the new tools were used mostly for coarse and heavy work. This defect has been, in some if not in all brands, completely overcome; and high-speed tools are now in use not only for the finest grades of metal cutting, but for wood cutting, which certainly is an extreme test of the smoothness of a cutting edge. It may be thought strange that high-speed steel should be used in a wood shop, where speeds have for a long time been as high as is expedient and safe. There is no particular gain in speed in wood cutting. The advantage in putting high-speed steel tools to this use lies almost wholly in their superior wearing quality. The immediate first cost usually is greater, but the life is incomparably longer, while the cost of maintenance is trifling. This of course is important where many such tools are used. Marvels of the New Tools. — The thing in the way of running ordinary tools at high speed when working on metals has already been shown to be the overheating of the tool near the cutting edge, and the con- sequent drawing of its temper, which is of course quickly followed by the rubbing away of the edge and the ruin of the tool. This series of steps follows as a result of the friction of the chip bending and sliding over the tool. The high-speed steels (are not, within certain limits, thus affected. Indeed, they seem almost to require abuse in order to bring out their highest capabilities. It is a common experience in shops that tools of these steels will not work to the best advantage until they have been run a little while and " warmed up." The speed capabilities and cutting power of these tools are indeed marvelous, compared with former experience. Thirty feet of chip per minute is a good performance with carbon tools; and the average on such work is not likely to be much over twenty, in a well-regulated shop. A hundred feet per minute has been mentioned as the extreme record of carbon-steel tools; and under ordinary conditions rarely is it possible to attain fifty feet con- tinuously. But high-speed tools in more than one shop cut a hundred and fifty feet and more as a regular performance, and higher speeds still are not at all unusual. Though this is several times as fast as was formerly possible, and is perhaps about twice the average in ordinary shop practice with the new tools, it is by no means the limit. It has been demonstrated that such tools can be worked up to more than three hundred feet per minute; and it is claimed that a speed of four hundred feet has been maintained continuously in cutting carbon steel with a comparatively small cut {\ inch) and a slight feed. Mr. J. M. Gledhill, a well-known authority on the subject, has asserted (Proceedings of the Iron and Steel Institute, New York meeting, October, 1904) that five hundred feet is attainable. At the Paris exposition of 1900, where high-speed tools were first publicly demonstrated, there was shown a lathe tool working for con- SELF-HARDENING AND HIGH-SPEED STEELS 19 siderable periods at a speed so great that the nose of the tool was red much of the time. In present regular practice tools are not permitted to get red hot, for softening to some degree inevitably takes place and the edge does not hold up. The chips as well as the tools of course get hot; and the former often come off with a deep-blue color when cut at a very high speed. The Chip Problem. — The removal of steel chips, when coming off so fast, is in itself a problem. In one test the services of two laborers were required to keep the machine clear. The total amount of metal removed in the case of some very large machines built with the purpose of utiliz- ing the new tools to their limit, is prodigious. Obviously there is nothing but disadvantage in removing metal unnecessarily; but there are cases where heavy cuts are unavoidable, and even economical. Under these circumstances it is not uncommon for cuttings to be removed at the rate of two thousand pounds per hour. Nor is this by any means the limit. There are recorded tests where double this rate has been at- tained for short periods. In actual everyday practice it is not unusual for a tool to cut several hundred pounds of chips per day, and that too without having to be ground more than once or twice. Inadequacy of Old Machine Types. — The pressure exerted in taking heavy cuts and the power required to drive machines at the high speed demanded by the new tools are tremendous. In a test, already men- tioned, the stress upon the tool was in excess of one hundred tons. To resist such forces'and to hold the tool and work firmly in proper relation, ordinary machines are quite inadequate. Machines a few years ago considered paragons of efficiency are, since the advent of the new steels, able to utilize but a fraction of the total efficiency of the tool; and while there is usually considerable advantage in the use of the new tools, it is only by using extremely heavy and rigid machines of the newer type that the full advantage is to be secured. Revolution in Machine Shop Practice. — Revolutions do not occur in a day, especially in the industrial world. There have been great indus- trial changes, several of them indeed, within the hundred and some years since handiwork ceased to be the chief agency in production. But these changes have been rather in the nature of evolutions. A new discovery or method gradually made itself necessary, and after a time there was a new order of things in which the former methods, once the standard of efficiency, became antiquated and had to be abandoned. High-speed steel evidently is one of those discoveries which will eventu- ally bring about a new regime in the metal-working industries. And while there is small likelihood that things will be very quickly upset, the carbon-steel tool made obsolete and the machine of yesterday's design antiquated, it now appears certain that after a few years carbon- steel tools will have a small place anywhere. Even razors are now 20 HIGH-SPEED STEEL sometimes made of high-speed steel and are said to hold an edge better than carbon steel. It is true that the new steels are as yet little known in many shops, especially the smaller ones, and that they are used very stingily in many large ones. The high cost seems to deter many from using them, and perhaps also in some cases the failure of first trials made without a satis- factory knowledge of how to make or use the new tools. Tungsten, molybdenum, and other like steel-hardening metals are rare, and con- sequently expensive, costing as high as $7 a pound in some cases. The manufacture of high-speed steel likewise is a more expensive pro- cess than that of making carbon steel. Of course for high-class tools crucible steel has been generally used; and crucible steel costs more to make than that produced by other processes. But even so, crucible steels sell as low as five cents, and occasionally even less, per pound; though the better grades, those most commonly used in tool making, sell anywhere from ten to twenty cents a pound, according to the grade; whereas ordinary high-speed steels sell for sixty to seventy-five cents per pound, and the " improved " steels run considerably higher in price. Efficiency of High-Speed Tools. — Eventually, no doubt, as processes are simplified, the cost will be much lower. Even at the present high cost, however, considering what it is capable of doing, high-speed steel is usually cheaper in the end. It has been a not uncommon experience that the cost of machine work on particular parts has been reduced one-half or more, though of course this could not be possible in many cases. If so, the industrial revolution in the metal trades undoubtedly would not only be at hand, but quickly accomplished. If the time required for cutting were all that entered into jobs of this sort, of course the cost reduction would be in proportion to the cutting speed. But every user of tools knows that often more time is required to get ready for a piece of work than for doing it. Frequently, indeed, the cutting time is an insignificant fraction of the total required; and obviously in such cases small economy of time could be expected. Nevertheless it is a rare case in which the tool-maintenance account is not capable of being reduced to a greater extent than the first cost is increased. Contrary to the general belief, the greatest saving effected by the use of high-speed steel ordinarily is not in the cutting of hard materials. The economy here is great, in general; but upon soft material it is prac- tically double what it is in the case of hard material. The Modern Way of Making Discoveries. — Many, perhaps most, of the important discoveries which have made civilization what it is were accidental, or at any rate not designed. Fortuity nowadays however plays a relatively unimportant part in discovery and invention. Even though an important idea be stumbled upon undesignedly, its perfection into a useful invention usually involves painstaking development. Men SELF-HARDENING AND HIGH-SPEED STEELS 21 conceive problems and set themselves earnestly to their solution, calling to their aid all available previous experience and knowledge in science or whatever may have bearing upon the problem in hand. It does not follow that the problem is always satisfactorily solved. Likewise it not infrequently happens that an investigation started to get at one thing, eventuates in something else not closely related. At some point in the development there may be indications pointing to divergent or even very different lines of investigation, which if carried on intelligently lead to results possibly more important than those first sought. Something like this was the development of the high-heat treatment of tungsten steels. The problem to which Mr. Taylor had set himself some twenty years before his most important discovery, was the development of a rational system of shop management — one that would obtain the highest possible efficiency in men and machines. Naturally, good tools, the best that could be made, were essential to such a system; and experi- mentation along this collateral line came to be perhaps the most impor- tant of all. In the pursuit of these investigations and the development of high-speed steel a very large amount of money was spent, many mistakes were made, and infinite patience was exercised. Something like fifty thousand recorded tests were made besides a great number not recorded, and close to a million pounds of steel and iron were cut into chips, the total expense having been estimated as not far from $200,000. Since the tungsten steels and their peculiar treatment have become generally known to the industrial world, and others have undertaken to experiment with them, probably much more has been spent in their further development. And still high-speed steel is but in its infancy. Like other inventions, it will undergo a process of evolution, one so complete, let it be hoped, as to leave little to be desired in respect of its utility. There is undoubtedly still plenty of room for painstaking and patient investigation and experimentation. In spite of all which, however, high-speed steel has already had a marked influence upon production, so that no shop can be said to be up-to-date which ignores its possibilities. CHAPTER III. NATURE AND CHARACTERISTICS OF THE NEW STEELS. Alloy Steels. — Until comparatively recent times the name steel was given by general consent to such a combination of iron and carbon (as we now know), together usually with slight proportions of certain other substances, as possessed the qualities of high tensile strength, homogeneity, toughness, and ability to resist crumbling, and which when treated in a particular way became considerably hardened. Later the distinction between steel and some varieties of iron became so slight that now it is very difficult to v make a definition which will in- clude, even in the case of the carbon steels, all those iron alloys commonly designated steel, and which at the same time will exclude those of practically identical composition, thoiigh perhaps of different structure, which are admittedly not steel. It is, in fact, impossible to draw a sharp line between mild steel, produced in an open-hearth furnace, and iron made by the puddling or other process, except for the presence of slag in puddle iron. The latter not infrequently has a higher carbon content than the mild steel. Before the development of the modern processes, it was comparatively easy to decide whether a given sample was steel or iron. If it hardened on being quenched in water after having been heated to a good red, it was plainly steel. But mild steel, with its low content of carbon, does not harden any more than wrought iron does. With the advent of the newer steels still greater difficulties are in the way of a suitable and precise definition, and it would be hazardous to venture one here. It is sufficient for our purpose to take for granted that the name steel may properly be applied to any alloy of iron with carbon, or of iron and carbon in combination with other of the so-called hardening elements, which permits hardening and temper- ing in a way to combine a relatively high tensile strength, reluctance to fracture, and resistance to crumbling. Nomenclature. — Since the discovery that substances other than carbon in virtue of their presence give iron the quality of becoming hard and tough under certain treatment, or at any rate assist carbon in producing this result, it has become necessary to make distinctions between the various kinds of steels, and it is now customary to speak of them in a general way as carbon, mushet or air-hardening, and high-speed or rapid steels. Various other terms have been suggested, but have not 22 NATURE AND CHARACTERISTICS OF THE NEW STEELS 23 come into general use, though the term "alloy steel is frequently used to designate all other steels than those depending upon carbon chiefly for their specific qualities. The alloy steels in turn are frequently designated as vanadium steel, tungsten steel, and the like, according to the distinguishing alloy; and because tungsten was the first and still is the most common of the elements used in the alloy steels, they are often spoken of as tungsten steels even though that element be in particular cases of minor importance or quite absent. The several alloy steels are used for various purposes to which their individual characteristics particularly fit them. Nickel steel, for instance, is largely used for armor plates and projectiles, and chrome and vanadium steels are largely used for the structural parts of machinery subjected to great strains, as in the case of certain automobile parts. It is not with this use of alloy steels, however, that we are at present concerned. Composition of Ordinary Steels. — Ordinary carbon steel, such as has through the ages been used for tools, contains small proportions of elements other than iron and carbon. Some of these are useful and perhaps even necessary to make the steel easily workable, either in forging or melting. This is the case of silicon and manganese. Both tend to make steel sound by preventing the formation of blowholes. Silicon, in the quantities usually present in tool steels, has small, if any, effect upon the tool; though in steels for some other purposes, where the proportion of silicon may be larger, it causes stiffness and possibly also adds to the hardness. When present in excess of say three or four per cent it causes brittleness and red shortness. Manganese acts as a sort of antidote for sulphur, phosphorus, and perhaps other impurities found in steel. It tends to prevent red shortness, promotes the forma- tion of fine and uniform crystallization, increases fluidity when the steel is melted, and makes it easy to work under the hammer or in rolls. Excess of manganese, however, makes steel cold short and causes sur- face cracking, especially upon quenching. Certain other elements, however, as phosphorus and sulphur, are not only useless but distinctly harmful; and the greater the proportion of either present, the more inferior the steel. Sulphur tends to make steel " red short " (brittle at a red heat) and therefore difficult to forge; while phosphorus tends to make it " cold short," and therefore brittle when cold. A very minute proportion of either will make a steel worthless for tools of almost any sort. Steel for cutting tools is usually expected to contain less than 0.02 per cent of either, though in some mushet or air-hardening steel the sulphur and phosphorus content have each been found to exceed 0.05 per cent. In extra special grades both sulphur and phos- phorus are kept below 0.008 per cent. The following table, giving the percentages of the various constituents of crucible steel intended for 24 HIGH-SPEED STEEL tools, indicates approximately the degree of purity required and the amount of carbon desirable in steel for the several purposes named. TABLE I. Man- Sili- Sul- Phos- Us:e. Iron. ganese. con. phur. phorus. Carbon. Hammers and other battering tools 99.040 0.21 0.21 0.022 0.020 0.50 to 0.75 Knives and shears, hot cutting 98.935 0.20 0.18 0.020 0.0150.65 toO. 80 Drills, reamers, dies, etc . 98.731 0.18 0.21 0.015 0.014 0.85 to 1.30 Lathe tools, knives, chisels, etc. 98.520 0.26 0.20 0.010 0.010 1.00 to 1.30 Razor steel . 98.265 0.22 0.20 0.006 0.009 1.30 to 1.50 Graving tools, etc. 98.374 0.16 0.14 0.014 0.0121. 30 to 1.50 Variations in Composition. — Of course the content of the various elements is not definitely established for tools intended for any par- ticular use. The practice of different steel makers varies, as also do the requirements of users, with respect to the composition of steels for special purposes. The above table therefore serves mainly to give some idea of the practical applications of the varying proportions of carbon and other elements. It will be seen that ordinary carbon tool steels, speaking in a general way, are constituted of iron, very small proportions of silicon and manganese in combination with carbon rang- ing from 0.5 per cent to 1.5 per cent, and minute quantities of impuri- ties such as phosphorus and sulphur. The variations possible in the carbon content of tools is well illustrated in the analyses of three well- known brands of carbon steel in use for lathe tools, whose performances were practically identical. The figures are quoted in part from Taylor. It is seen that the percentage of carbon varies from 0.681 to 1.240. TABLE II. Steel. Iron. Manga- nese. Silicon. Sulphur. Phosphorus. Carbon. Tungsten. Ill and Z II S 98.524 98.350 98.867 0.189 0.156 0.198 0.206 0.232 0.219 0.017 0.006 0.011 0.017 0.016 0.024 1.047 1.240 0.681 0.079 Constituents of Self-Hardening Steels. — Besides carbon, manganese and silicon, self-hardening steel contains a considerable proportion of tung- sten, chromium, molybdenum, vanadium, or certain other like elements, generally in definite combinations, as hereafter mentioned. The silicon content is practically the same as in carbon steel, while -the manganese is usually considerably higher, varying from rather more than one per cent to above three per cent according as the tungsten is high or low. High carbon also has been the rule in these steels, the percentage running say from somewhat more than one, to two per cent, and even higher, although the present tendency is toward reducing the carbon content NATURE AND CHARACTERISTICS OF THE NEW STEELS 25 in high-speed steels, and it is occasionally found very much lower than one per cent. Chromium, when present, takes the place of a portion of the manganese or the tungsten, which latter ranges from about 4 to 11 or 12 per cent. The analyses here given are characteristics of these steels. TABLE III. Steel. Carbon. Tungsten. Molybde- num. Chromium. Manga- nese. Silicon. Sul- phur. Phos- phorus. Mushet Midvale B C D 2.150 1.140 1.615 1.750 1.842 5.441 7.723 10.000 11.589 4.580 0.398 1.830 3.430 1.000 2.694 1.578 0.180 1.650 1.750 2.430 1.044 0.246 0.285 0.060 0.890 0.016 0.007 0.027 0.023 E 1.220 7.020 0.078 0.300 0.180 0.010 0.017 A Singular Anomaly. — From the fact of its containing rather more than 7 per cent of tungsten, the steel marked E would naturally be thought self-hardening, like the others. This however is not the case. Though it has a tungsten content about a half greater than that of some self-hardening steels, this steel has no such property, when heated to the customary temperatures, at any rate. It hardens only on being quenched in water, as is the case with ordinary carbon steel. This circumstance naturally raises the question as to what causes tungsten steel to be self-hard, and likewise that of why steel of any kind becomes hard under certain conditions. Theory of Steel Hardening. — The' hardening of ordinary carbon steel, as is very well known, is accomplished by heating the piece intended to be hardened to a red color ranging between a dark and bright cherry (something like 735 degrees C. or 1350 F.), and then quenching it in a water or other suitable bath to about the normal temperature of the air. This process seems to change entirely the structure of the steel as seen under the microscope. Careful investigations into the nature of these changes have been made, and a number of theories or hypothe- ses have been advanced to account for, or rather to explain them. The several hypotheses differ more or less among themselves; but those that are most generally received agree substantially that steel may exist, according to temperature or quenching temperature, in three type forms. At temperatures below 735 degrees C. or thereabouts, carbon steel is in the unhardened or annealed state. Between 735 degrees C. (1350 F.) and 820 degrees C. (1510 F.) it exists in a hardened state; and above 820 degrees C. it exists in a state harder than the first and softer than the second, and is at the same time very tough. The Constitution of Annealed Steel. — Steel is not a simple substance but, as shown by the microscope, is a conglomerate of crystals of dif- 26 HIGH-SPEED STEEL Fig. 8. Typical structure of annealed Fig. 9. Structure of hardened high- steels. X150. speed steel. X 1,000. From Mr. C. A. Edwards' paper on "The Function of Chromium and Tungsten in High-Speed Steel, in the Journal of the Iron & Steel Institute. Fig. 10. Microscopic structure of high-speed Fig. 11. Microscopic structure of high-speed steel ingot as cast. X 150. steel ingot as cast. X 1,000. From Dr. Carpenter's "Possible Methods of Improving Modern High-speed Turning Tools." 3 4 Percentage of Carbon Fig. 12. Constituents of annealed carbon steels. NATURE AND CHARACTERISTICS OF THE NEW STEELS 27 ferent substances. For example, annealed steel containing 0.90 per cent carbon, consists entirely of crystals having a pearly appearance under the microscope, to which the name of pearlite has been given; while in steel of less than 0.90 per cent carbon the microscope shows pearlite with varying amounts of a substance called ferrite, the pro- portion of which increases as the amount of carbon decreases from 0.90 per cent. Steel of more than 0.90 per cent carbon consists of pearlite again with varying amounts of silvery white crystals of cemen- tite. The properties of a steel depend upon the properties of pearlite, ferrite and cementite, and upon the proportions in which these substances exist in it. Graphical Illustration. — The amount of pearlite, ferrite or cementite in different compounds of iron and carbon is shown graphically in Fig. 12, in which the distance from the axis 00' will represent the amount of carbon in the material, and the vertical distance above the axis HH' in the different areas there shown, will represent the percentages of pearlite, ferrite, or cementite present. For example, a line drawn from A to A', which is at a distance from axis 00' corresponding to 0.90 per cent carbon, will be entirely in the pearlite area, and will show that this steel contains 100 per cent of pearlite, as before stated. Steel represented by the line drawn from B to B' , which is halfway between AA' and 00' , will contain 0.45 per cent carbon, and will consist of 50 per cent pearlite and 50 per cent ferrite. The line drawn from C to C" will represent material containing 3.75 per cent carbon, and will lie one-half in the pearlite area and one-half in the cementite area, showing, therefore, 50 per cent pearlite and 50 per cent cementite. The material represented by the line 00' will contain no carbon and consist entirely of ferrite, and the material represented by the line PP' will contain 6.6 per cent of carbon and consist entirely of cementite. By a similar method we can determine from the chart (Fig. 12) the proportion of pearlite, ferrite, or cementite in steel of any percentage of carbon, it being understood that combined carbon only is considered here, free carbon, which is what we call graphite, not being a normal constituent of ordinary steel. Knowing the properties of pearlite, ferrite and cemen- tite, and determining from the chart the proportion of each in steel of any given carbon, we can estimate from these data something of the characteristics of the steel in question when it is in the annealed condition. Properties of Pearlite. — Steel consisting entirely of pearlite has the finest crystalline structure of any carbon steel, and this is accompanied by the greatest strength and a high degree of hardness. When annealed, this steel also has a degree of toughness, so that it can be bent double while cold, and a wire of about \ inch diameter can even be tied cold in a knot without cracking. Furthermore, it is capable of receiving, by hardening or tempering, the greatest possible combination in a carbon steel of 28 HIGH-SPEED STEEL two valuable properties for cutting work, viz., hardness with absence of brittleness. True, tempered steel with less than 0.90 per . cent carbon will not be so brittle as pure pearlite steel, but, on the other hand, it will not be so hard, either; and tempered steel with more than 0.90 per cent carbon will be harder than pure pearlite steel, but will also be more brittle. Properties of Ferrite. — Ferrite crystals contain theoretically no carbon or other impurities; in other words, they consist of pure iron. There is no material sold commercially corresponding to pure ferrite, but the purest forms of irons, such, as Swedish wrought iron, electrolytic iron, Fig. Both X 1,000. 13. From Ferrite. Fig. 14. Pearlite (lamellar). Dr. H. C. H. Carpenter's paper before the Iron and Steel Institute. Fig. 15. Pearlite (black) segregated in ferrite. Fig. 16. Gementite (black) with contained . . patches of ferrite. Dr. H. C. H. Carpenter. Both X 250. and the most refined products of the electric smelting furnace, come the nearest to it. Those who are familiar with Swedish wrought iron can therefore judge of the properties of ferrite, and need not be told that they comprise great toughness, softness and ductility, together with high electric conductivity and magnetic strength. Crystals of ferrite in cutting tools therefore will decrease their ability to cut, but at the same time increase their toughness. Properties of Cementite. — Cementite crystals contain 6.6 per cent carbon and are a chemical compound of iron and carbon, known as iron carbide, and having the chemical formula Fe 3 C. Pure cementite NATURE AND CHARACTERISTICS OF THE NEW STEELS 29 does not occur commercially, but its crystals can be separated chemi- cally from high-carbon steel, and its properties studied from them. The characteristics of cementite are its brittleness, lack of strength, and great hardness, which latter is not increased by the ordinary processes of hardening or tempering, and cannot be decreased by annealing because annealing of pure cementite breaks up the chemical compound and converts it into a substance similar to annealed malleable cast iron. Even when cementite is mixed with a large proportion of pearlite, annealing or even heating to a bright red will break up the compound to some extent and precipitate some of the carbon, and this is the source of specks of graphite occasionally found in steels containing free cementite, that is, in steels with 1.50 per cent carbon and more. Effect of Heat Treatment on Steel. — The constituents mentioned above are those normally found in steels which have cooled slowly from the temperatures at which they were cast or rolled, or in steels which have been annealed. All of these constituents are changed by heating to higher temperatures, and this is the effect of what is called heat treat- ment, viz., hardening and tempering. Changes Occurring on Heating Pearlite. — When steel consisting en- tirely of pearlite is heated to about 735 degrees C. (1355 degrees F.), it undergoes a change in structure and properties. The pearly appear- ance under the microscope is lost and we see a homogeneous white substance made up of polyhedral crystals to which the name of austenite has been given. At the same time the molecule of steel becomes hard and loses the power of being attracted by the magnet, so that at this high temperature, and above it, steel is non-magnetic so far as the ordinary test shows. When the steel is again cooled slowly below this temperature, the austenitic structure changes back to pearlite, the molecule loses its hardness, and the magnetic property reappears. Effect of Heating and Cooling of Pearlite with Ferrite, or of Pearlite with Cementite. — Most of the steels used for cutting tools are, by composition, chiefly made up of pearlite. Even if there be some fer- rite or some cementite present besides the pearlite, the changes occur- ring in the steel on heating and cooling have ultimately the same effect as that just described, although they occur in a somewhat dif- ferent way, which will be explained later. For the present, therefore, the changes in all carbon steels will be considered as if they were merely from pearlite to austenite and austenite back into pearlite. When steel with less than 0.90 per cent carbon is heated to about 735 degrees C. (1355 degrees F.), all the pearlite in the steel changes, as before de- scribed, into austenite, but the ferrite remains as ferrite for the time being. If the heating continues, however, ferrite is gradually absorbed by the austenite with each rise in temperature. When there is as much as 25 per cent of ferrite (that is, if the steel contains 0.67 per cent car- 30 HIGH-SPEED STEEL bon), the ferrite is not all absorbed until a temperature of about 770 degrees C. (1415 degrees F.) is reached. When the steel consists of 50 per cent pearlite and 50 per cent ferrite, the ferrite is not all absorbed until a temperature of nearly 800 degrees C. (1462 degrees F.) is reached. However, even when there is as much as 99 per cent of excess ferrite, this ferrite is all absorbed and the mass converted into austenite by the time the temperature has risen to 935 degrees C. (1715 degrees F.). If we have cementite present in excess of the pearlite, instead of excess ferrite, this cementite is absorbed into the austenite somewhat more slowly than the ferrite was. Even with only 15 per cent of excess cementite (steel of 1.90 per cent carbon) the temperature will rise to the point where melting begins (1150 degrees C. or 2102 degrees F.) before the last of this cementite is absorbed. With 9 per cent excess cementite (steel of 1.50 per cent carbon) the cementite is all absorbed by the time the steel has reached 925 degrees C. (1697 degrees F.). It is evident now that all steels are converted completely into the non-magnetic austenite, with the hard molecule, upon heating to a sufficiently high temperature, and it will not generally be necessary hereafter to distinguish in this regard between steel consisting of pure pearlite and that consisting of pearlite with cementite or of pearlite with ferrite, since the only effect of the excess substance is to raise the tem- perature at which the conversion becomes complete. The same relation holds good in the cooling of these steels. If there is excess cementite or excess ferrite present, as the case may be, the excess is first precipi- tated at a somewhat higher temperature than the change from austenite to pearlite, depending upon the amount present, and finally, when all this excess is precipitated, the residual austenite is converted into pearlite. It is to be remembered, however, that if the carbon is below 0.90 per cent, ferrite is later absorbed into the austenite on heating and precipitated in advance on cooling, while, if the carbon is above 0.90 per cent, cementite is so absorbed and precipitated. Lag. — One characteristic of the change from pearlite to austenite on heating, and the reverse change from austenite back into pearlite on cooling, deserves special notice; that is, it is a somewhat tardy one and does not take place instantaneously. On the contrary, it requires a few moments for its completion. We express this by stating that the change lags behind the temperature to some extent, and this tardiness, or lag, is greater the more rapid the heating or cooling is. We may liken it to the tail of a comet. If the comet is traveling through space at a very rapid pace, its tail will drag out behind it for a long distance, but, if the comet is traveling with a relatively low speed, the tail may almost keep pace with it, and may even surround it. So, if we heat pure pearlite steel very slowly, so that it may take many hours, or even days, to reach a bright red heat, the change from pearlite to austenite may NATURE AND CHARACTERISTICS OF THE NEW STEELS 31 occur all at approximately the same temperature, and if then the steel be cooled equally slowly, the change from austenite back into pearlite may also be completed all at the same temperature. It has been shown that under these conditions of very slow heating and cooling, the change from pearlite to austenite on heating will occur at practically the same temperature as the reversion from austenite to pearlite occurs on cooling, although, with the ordinary rate of heating, the change from pearlite to austenite is not completed until a temperature of about 735 degrees C. (1355 F.) is reached, and with the ordinary slow cooling in the furnace 000 375 850 825 800 775 750 725 700 675 650 025 600 V A=s 8.2°C 1 V '7 764.2' C ^ - 726.2°G £/ = 707 .4°C^ -<^=708. 6 C \ t 7 s a ' = 654.4° C \& P A & k i"a ) ' i 2 3 4 5 6 7 J 8 9 ) 10 ^ w FlQ. 17. Curves showing relative location and nature of critical ranges in high-speed and in carbon steels. The lower position of the cooling recalescence range, compared with the heating range, is shown also. the reversion from austenite to pearlite does not occur until the tem- perature has dropped to 690 degrees C. (1274 F.). This difference in tem- perature on heating and cooling is not because the change is not a truly reversible one, but merely because it is a tardy change and lags behind the temperature in both directions. To return to our simile of the comet: If we can imagine a comet traveling with great speed from west to east,_with its tail extending a long distance behind, it would have to pass well beyond a given point before the last of its tail would reach that point. If now we should suddenly reverse the direction of the comet, the momentum of the tail would still carry it on and the comet itself would have had to pass some distance beyond the given point in 32 HIGH-SPEED STEEL the opposite direction before it would have again dragged the last of its tail beyond that point. We might even imagine the return of the comet to be so extremely rapid that it would become separated from a part of its tail, and leave it behind, beyond the given point we have been considering. Theory of the Hardening of Steel by Sudden Cooling. — If any carbon steel be heated to a high enough temperature, it will be entirely con- verted into austenite. If now the steel in this condition be cooled with great rapidity, we may bring it to the atmospheric temperature with a part of its molecules still left in the austenitic condition; and since the austenitic molecule is much harder than the pearlitic molecule, it is on No.l No.2 No.3 ]\ T 0.4 No.5 Fig. 18. Gradual reduction and final disappearance of marked variations in the curve (apparent dis- appearance of recalescence points) on cooling from increasingly higher temperatures. From Dr. H. C. H. Carpenter's paper, "Possible Methods of Improving Modern High-Speed Turning Tools." this principle that the hardening of steel by sudden cooling is explained. No cooling has ever been so rapid, however, as to obtain all the molecules in the austenitic condition, because, although the change from austenite to pearlite is a tardy one, it is not so slow that we can get away from it altogether. Influence of Carbon on the Hardening. — There are some conditions, however, which aid in keeping some of the steel in the austenitic con- dition. One of these is the influence of carbon, which tends to obstruct the change and make it more tardy. Thus, with 1.60 per cent carbon in steel, and with very rapid cooling, we may find under the microscope as much as 70 per cent of the mass in the austenitic condition, but this NATURE AND CHARACTERISTICS OF THE NEW STEELS 33 is about the maximum that has been obtained up to this time by quick cooling and carbon content only. Again, the importance of carbon in this respect is shown by the obser- vation that, unless some carbon is present, we cannot retain any of the steel in the austenitic condition, no matter how rapid the cooling or what other elements — tungsten, manganese, chromium, etc. — be pres- ent. We may cool it with such extreme rapidity that it is brought to the temperature of the atmosphere in the austenitic condition, but with no more than traces of carbon present it gradually changes over, even when cold, to the pearlite condition. We may express this by saying that we can catch, or trap, the austenite in the steel by means of quick cooling, but we cannot " fix " any of it there except with an Fig. 19. Martensite. X 150. Fig. 20. Martensite and cementite completely separated. X 1,000. influential proportion of carbon. The tendency of austenite to revert to the normal, or pearlitic, condition at all temperatures below 700 degrees C. (1292 F.) is so strong that it will do so even when cold unless carbon is present as a fixing agent. The colder the steel, the more slowly the reversion proceeds, however. Martensite. — In order to understand the nature of high-speed steels and the reasons why tungsten and other elements, together with the special heat treatment required, produce the effect they do, we must recognize another constituent of steel which is intermediate between pearlite and austenite, and to which the name of martensite has been given. Martensite is probably never a normal constituent of steel, but occurs only as a transition stage in the change from austenite 1 to pearlite. 1 Alpha, Beta, and Gamma Forms of Iron. — Although a knowledge of the nature of austenite, martensite, troostite, and pearlite is not necessary to an understanding of high-speed steels, there are many no doubt who wish to know as much about the tools as it is possible to learn. In order for us to understand the distinction between these different constituents in hardened and tempered steels, we must first understand the varying and compound nature of iron itself. Pure iron at atmospheric temperatures is the most magnetic substance known to man, with the exception of a recently discovered silicon steel alloy, which it is not necessary for us to discuss here. When this pure iron is heated to and above a tem- 34 HIGH-SPEED STEEL ap-eaSijuao saa^Saa ' aaniviactawj) c _ >> 3 * '^ o c s «j . c c 8 -3 5 £ £ O o fi » M 1- v-t o e O OQ (2 P4 S w 1 5 s a- e e 5 < 5 5 e c c > > 3 1£ i i •* o apBJSyuao saajSaa 'sxtijeasiluisi NATURE AND CHARACTERISTICS OF THE NEW STEELS 35 That is to say, austenite changes first into martensite, but, as this has no normal place in steel at any temperature, it should change at once into pearlite. It is to be observed, however, that the change from austenite to martensite is a very quick one and most difficult to prevent or obstruct, while the change from martensite to pearlite is a much slower one. Martensite is harder and more brittle than austenite, and also more, bulky; so that the steel expands when it changes from austen- ite to martensite, and then contracts again when its proceeds further to the pearlite stage. This expansion explains the occasional bursting of steel tools during hardening. Fixing the Austenite and the Martensite.- — The usual heat treatment for hardening tools, which consists in raising them to a bright red heat and then plunging into cold water, is not quick enough to prevent the change from austenite to martensite. Austenite can only be fixed at atmospheric temperatures when (1) the cooling begins from a strong white heat, when (2) it is the most rapid possible, and when (3) the carbon is over 1 per cent. The method of cooling usually employed for this extreme rapidity is to plunge the white hot steel into iced brine or some other liquid cooled well below the freezing point of water. FiG< 23- Austenite . x 90a Even under these extreme conditions only a part of the austenite can be preserved unaltered. The usual process of hardening produces martensite in the steel, whose hardness, as has been stated, is greater than that of austenite. Unfortunately, however, this hardness is. accompanied by too much brit- tleness to withstand the shocks of service, and generally some of it has to be sacrificed for the sake of durability. This sacrifice is made by means of a tempering process. Tempering. — It has been already said that the colder the steel, the slower the change from austenite to pearlite, after it has been trapped by a hardening process. Advantage is taken of this fact in the com- perature of 760 degrees C. (1400 F.), it loses almost completely the power of being attracted by a magnet. At the same time the iron changes its electrical conductivity, its form of crystallization, and other properties. If the heating be continued, the form of crystallization and other properties change again at a temperature of 890 degrees C. (1634 F.). In short, pure iron can radically change some of its properties at different temperatures without changing its composition or individuality. Because the iron remains iron all the time, we therefore call these different forms in which it occurs "allotropic modifications." That condition in which it exists at atmospheric temperatures we call Alpha iron (a); that at temperatures between 760 and 890 degrees C. (1400 and 1634 F.) we call Beta iron (/?); and that above 890 degrees C. we call Gamma iron (7), 36 HIGH-SPEED STEEL FrG. 24. Curves showing clearly the lower critical point — that connected with the phenomena of tempering. Movements of the differential galvanometer, about fa size. Dr. H. C. H. Carpenter, Journal of the Iron and Steel Institute. Carbon Self -hardening High Speed Tool Tool Tool No.l No.2 No.3 Fig. 25. Curves comparing softening or tempering ranges of carbon, self-hardening, and high-speed steels. From Dr. Carpenter's paper, "Possible Methods of Improving Modern High-Speed Turning Tools." NATURE AND CHARACTERISTICS OF THE NEW STEELS 37 mon process of tempering. If we have fixed some martensite in a carbon- steel tool, we can " let it down " toward the pearlite stage to any desired degree by gently heating the tool. Softening begins by this process at a temperature of about 200 degrees C. (392 F.). This proceeds more and more as the temperature is raised, and usually is as complete as desired before we reach a temperature of 400 degrees C. (752 F.). All temper- ing after hardening will therefore take place at the temperatures between those mentioned, and in fact 99 per cent of all tempering of carbon-steel cutting tools is between 200 and 300 degrees C. (392 and 572 F.), while it is only those comparatively few tempered articles whose hardness must be plentifully sacrificed for the sake of toughness, such as springs, screw drivers, cold chisels, etc., in which we let down-the martensite by tempering above 275 degrees C. (527 F.). Temper Colors. — Nature has fortunately provided a tolerably accurate and convenient pyrometer whereby the extent of tempering may be esti- mated by eye; for, at about 200 degrees C. (392 F.) the steel assumes a Fig Fig. 27. Fig. 28. Heated to 680 degrees C. X 1,000. Fig. 29. Heated to 730 degrees C. X 1,000. Change in microscopic appearance of high-speed steel due to "low heat" treatment. Composition of steel shown in Figures 36 and 37, C 0.68, Cr 3.01, W (tungsten) 19.37; that shown in Figures 34 and 35, C 0.67, Cr 6.18, W 12.5 per cent. very light lemon color. As the temperature rises, the color deepens to a faint yellow and then to a straw, pink, light purple, and so on, to a deep 38 HIGH-SPEED STEEL blue. The approximate correspondence of these colors with the differ- ent temperatures is shown in the frontispiece of this book. The temper colors are due to a film of iron oxide which forms on the bright surface of the steel and which gets thicker and thicker as the heat progresses. Tempering in Oil. — Steel is sometimes tempered, not by the com- bination process just mentioned, whereby it is first brought to the marten- siti'c stage by hardening and then let down towards the pearlite stage by warming until the desired " temper " is obtained, but by cooling with intermediate rapidity in the first instance, such as by plunging into oil when at a bright red heat, instead of into water. In this way the martensitic stage is not so completely fixed, and the moderately rapid cooling produces the same effect as if we first cooled with greater speed and then let down the martensite by tempering. Troostite. — Tempered steel consists wholly or partly of troostite, which is a transition stage between martensite and pearlite. Troostite is softer and tougher than martensite, so that the toughness of tempered steel will depend upon the relative proportion of troostite in it, and this latter will depend in turn upon the amount of tempering. Some troos- tite is found even in hardened steel unless it has been quenched from a temperature nearly a white heat and far above the " critical point," i.e. the point at which pearlite changes to austenite on heating, and austenite reverts to pearlite on slow cooling. Nature of Austenite, Martensite, Troostite, and Pearlite. — In the present incomplete state of our knowledge of the constituents of hardened and tempered steels, we must accept all theories ? somewhat tentatively. However, the indications at present are that austenite is a solid solution of carbon 2 in gamma iron, that martensite is a solid solution of carbon in beta iron, and that troostite is a solid solution of carbon in alpha iron. Theoretically neither beta nor alpha iron can be maintained in solid solution with carbon, and therefore austenite is the only stable one of these solutions, and martensite and troostite must be considered as unstable constituents and merely transition stages between austenite and pearlite. In other words, when the solution of gamma iron falls to the temperature at which it breaks up, it changes to the solution of beta iron; this being an abnormal solution at once breaks down into the solution of alpha iron (provided of course that it is not obstructed), and the alpha solution being also abnormal breaks down into pearlite, which is not a solution at all but a mere mixture of crystals of ferrite 1 The theory of steel hardening here outlined is substantially that proposed by Pro- fessor Howe, which is perhaps as widely accepted as any other. 2 Where we speak of solutions of carbon in gamma, beta, or alpha iron, it is to be observed that the carbon may be dissolved directly in the iron, Or it may be in the form of iron carbide, or cementite, and this dissolved in the iron. NATURE AND CHARACTERISTICS OF THE NEW STEELS 39 -and cementite so fine in structure that the high powers of the micro- scope are necessary to show its structure. Effect of Manganeseon Hardening. — Manganese has the effect of delaying the change from austenite to pearlite, and of acting as a fixing agent for the austenite, but both of these influences are different in kind from the influence of carbon. Carbon hinders the change by making it slower. Manganese, on the other hand, makes the change occur at a lower temperature. That is to say, instead of the change on heating and cooling taking place at about 700 degrees C. (1292 F.) it occurs 100 or so degrees lower when there is 1 per cent of manganese present. With 2 per cent of manganese, the change occurs at a lower tempera- ture still, and finally when there is as much as 12 to 20 per cent of manganese present, it occurs at a temperature below that of our atmos- phere. In other words, as 12 per cent manganese steel cools from the temperature at which it is made, it ordinarily never gets so cold that it will change over from the austenitic to the pearlitic condition. There- fore, manganese steel, which, as usually made, contains between 12 and 15 per cent of manganese, is what is called a self-hardening steel, mean- ing that it is normally in the austenitic or martensitic condition, and that even annealing will not change it to the pearlitic condition. Heat Treatment of Manganese Steel. — As has been shown, the change from austenite to martensite is a very quick one, and difficult to prevent. Thus even manganese steel, if allowed to cool slowly in the mold into which it is poured, will be wholly or partly in the martensitic stage at atmospheric temperatures. In this condition it is not only very hard, but also very brittle. By heating it to a white heat (over 1000 degrees C. or 1832 F.), and cooling very rapidly, as by plunging it into ice-water, we can, however, entirely fix the austenitic stage in this steel. This fixation of the austenitic stage makes the steel not so hard as it was when in the martensitic condition, but still very much harder than in the pearlitic. It is also very tough; but unfortunately it has such a low elastic limit that the thin edge of a tool will not stand up under the shocks of service, but crumbles away, so that manganese steel is not suitable for making cutting tools. Effect of Nickel on Hardening. — The effect of nickel on hardening is the same in kind as that of manganese, but it takes about twice as much nickel to produce this effect. As ordinary nickel steel contains usually only 3J per cent of nickel, and as it requires about 25 per cent to bring the temperature of the change below that of the atmosphere, commercial nickel steel is not a self-hardening steel, but is used for other purposes. Effect of Chromium on Hardening. — The effect of chromium on harden- ing appears to be similar to that of carbon in so far as making the change more slow is concerned. With 1 to 2 per cent of chromium in steel and about 1 per cent of carbon, we can get a much more intense degree of 40 HIGH-SPEED STEEL hardness by means of rapid cooling, but not without. That is, the steel is not self-hardening. Although chromium alone will not make a steel self-hardening, yet chromium with a few per cent of manganese (much less than that present in manganese steel) will produce this result. Chromium also has the effect of increasing the elastic limit of steel, especially when it is combined with vanadium. The hardness imparted by chromium is not accompanied by as much brittleness as that induced by carbon. When the steel contains chromium, the amount of carbon is reduced, since the extreme of hardness is not desired, and toughness may be gained thereby. Tools below 1.50 per cent of chromium are very 0.39 0.41 0.77 Percentage of Carbon 0.86 0.71 1.27 fl ■ ol jJjHe'el p* , k <9 & ,l'K d-* > 't > , '/ \ s — I — 1 — >— LTmiT "~ - J 1 \ \ V 5 «, # le£fr e ; "" - 1 E a.s ic Limit j! Am eaied Stee ^..- - V „ "., , 1 2 3 4 5 6 7 a 9 10 11 12 13 14 Percentage of Chromium Fig. 30. Effect of chromium on tensile strength and elastic limit. From "Steel and Its Uses," by Edmund F. Lake. tough and effective in cutting soft materials. High-chromium tools, containing up to 5 or 6 per cent chromium, are very hard and effective upon refractory materials. Mushet Steel. — Chromium with tungsten will also produce a self- hardening steel, although neither of these elements alone will have any such effect. The combination of tungsten with a small amount of manganese will also reduce the temperature of change below that of the atmosphere. The famous old mushet steels, which were the first self- hardening steels, were of this composition and character. Without either a little chromium or a little manganese, however, no amount of tungsten would produce such an effect. The anomaly exhibited by the last steel mentioned in Table III is now explained. That steel con- tains more than 7 per cent of tungsten, but only 0.3 per cent of manga- nese and little more than a trace of chromium, neither of which is sufficient in amount to make the steel self-hardening. Effect of Tungsten. — Tungsten acts first as a strong obstruction to all the steps in the change from austenite to pearlite, so that when we have 7 per cent or more of tungsten present, a moderately rapid cooling, even such as allowing the bar to cool in the air, will prevent the change to pearlite. Indeed, when tungsten steels are to be annealed and the NATURE AND CHARACTERISTICS OF THE NEW STEELS 41 pearlite stage produced, it is necessary that the cooling shall be very slow indeed, occupying several times as long as the annealing of normal carbon steels. Tungsten acts secondly as a powerful fixing agent for martensite. It has been already shown that martensite is not stable in normal carbon steel even after it has been induced there by hardening, unless the metal be kept co'ol. Warming it up to the so-called temper heats changes the martensite over to troostite, and if the heating be continued, even the effective hardness of troostite is lost long before the steel reaches a red heat. The presence of 7 per cent or more of tungsten, however, increases the stability of martensite so much that the steel may be heated well above the tempering heats before the martensite even begins to break down into troostite or pearlite. Red Hardness. — Red hardness is the quality of hardness when at a red heat, and tungsten imparts this to steel under certain conditions* for a c. Heating Curves of High-Speed Steels 1000° 1 "be C 900° 15 '** OS £ a r 1 a) / r Centigrade 8 o .hkely *£ **^ and are very deep in proportion ,n section, raiely exceeding 4 y ^ together by t0 t^ir cross secrtonUualy they « ^ ^ . t'olv^he ingot S surface and prevent its sticking to the sides of th T^mL -The slag having been skimmed off, the teemer quickly JpTs'the Juelbfelnto th/mold in pre^dy the ^manner - is done in teeming »^ «* g J^TstiU to teem properly, of the mold. The stream must not at any time in TnlY moMs'used in the manufacture of high-speed steel usually Ingot molds useam oruci ble though sometimes larger ones hold the contents of but a single crucible, inoj * DurD0Ses In this having been mixed in a ladle. the mold is ^o^o "^ed" S-f£ -pV Loken off, to remove the pipedTdsegregated P portion likely to be our ^^J^™ face of the remaining portion^ ^^f^ ^ out . At and such minor superficial ones as are msc satisfactory, the same time a sample ^ " ^nt IT — Ahe rejection and no physical detect Being iounu v,»«tinir not infrequently and remelting of the ingot, after a prolong^ heating not n q y lasting two days (for the ingots having been in chil m ,^ intensely hard) at a temperature close to » <"• P in the hammer and thoroughly worked out mto » rollg for thoroughly inspected, and if pert ect go * theb«nm» hammered finishing to the require [section _ Most high spe hamm6ring and nearly to shape, and then rolled to a nmsn. 60 HIGH-SPEED STEEL Fig. 44. Billets ooming from the shears. Fig. 45. Forging out the bars under the steam hammer. THE PROCESS OF MAKING HIGH-SPEED STEELS 61 rolling must be done at a heat considerably higher than that used in the case of ordinary steels. These alloy steels are so dense that they work well only when at a bright red or even higher temperature. If ham- mered or rojled at a lower temperature the metal does not flow freely and uniformly under the blows or pressure, and strains, if not cracks, are caused — generally bad cracks, which unfit the steel for its special use. Even if no cracks develop immediately, the strains thus set up, unless let down by subsequent thorough annealing, frequently produce cracks and failures long after the bars have been passed as perfect, and more than likely after they have been put to use. To insure freedom from defects the forging temperature is customarily near 1,000 degrees C. (1850 F.). Annealing. — High-speed steel being partially self-hardening, the bars when finished are hard and require annealing except for a very few purposes. Unless annealed the bars cannot well be used even when they only require to be ground and inserted in a holder, owing to the" difficulty of breaking off what may be wanted; and it is utterly impos- sible to machine high-speed steel in this condition into any of the many special forms required. The hard bars can of course be forged; but even in this case it is much better to use annealed stock, for several rea- sons, the most important of which is that annealing relieves the internal •strains set up in hammering or rolling and decreases the liability to future flaws or cracks. Also, the structure of the steel becomes uniform, homogeneous, and tenacious; and according to the testimony of some, its life is increased. The annealing is done in muffle ovens of the customary type, the heat being gradually brought up to a red heat of 800 degrees C. (1500 F.) or somewhat higher. The bars may then be removed and slowly cooled, but much better results are obtained when they are allowed to cool in the oven itself, the heat being shut off soon after the desired temperature has been reached, and the oven allowed to cool slowly. As in the other processes, great care must be taken that conditions are just right, else there is likelihood of the steel coming out poor or indifferent in quality, even when the mixture is good. Twelve to eighteen hours, according to the size and shape of the bars, is required for proper annealing. Refinement of Methods. — Those familiar with the process of making ordinary crucible steel doubtless will have noted already that the process of making high-speed steel differs from it in few important respects. In general the equipment used and the methods practiced are identical. The chief difference is in the stock put into the crucible, and in the exceeding care exercised throughout in producing the high-speed steel. In a mill which endeavors to make and keep up a reputation for pro- ducing a superior quality of high-speed steel, the extent and frequency 62 HIGH-SPEED STEEL of the examinations and tests of stock in process of manufacture, is surprising. The ingots are carefully analysed, and are inspected be- fore as well as after " topping." The topping itself is intended to remove any possible inferior metal or defects, frequently found in that portion of the ingot. The billet is again inspected; and each separate bar likewise undergoes examination for defects. The bars are generally "pickled" to make defects more easily discernible, if any exist. If defects appear at any time in the course of all these inspections, the material is rejected if inferior in quality, or re-melted if merely defective in structure. Considering the care necessary in its manufacture, and the high skill required in the workmen, it is not at all singular that high-speed steel continues to sell at the extraordinarily high price which it commands in the market. There are, however, additional reasons for its high price, chief among which is the high cost of the special alloying elements. Kind and Quality of Constituent Materials. — The materials used are necessarily of the purest, and certain of them are rare; for both of which reasons their cost is very high. Some makers use only the purest Swedish and Dannemora iron, saying that these alone are free enough from sulphur, phosphorus, and other impurities, to give the best results in high-speed steel. These irons are considerably more costly than even the best of ordinary kinds. A number of makers however, utilize good qualities of native muck bar, saying that these give results as good as can be obtained; but these extra pure irons also have a higher price than the ordinary. The tungsten, molybdenum, and other hard- ening metals, vanadium especially, are rare, their ores being found in but few places and those usually not easily accessible. These ores are commonly reduced in the electric furnace, sometimes to the metal- lic state, and at others to the ferro alloys, either of which can be used in the manufacture of high-speed steel. The prices of these metals range, in either state, from $0. 40 to $7.00 per pound. 1 Since the proportion of 1 The prices quoted in March, 1908, were approximately as follows: Tungsten $0 . 75 per lb. Vanadium 6 . 00 " Molybdenum 1 . 50 " Titanium 1 . 00 " Chromium 37J to .75 " The figures given are for the contained metal in the ferro state. Swedish iron was at the same time quoted at about three cents per pound. Ten years before these metals were stated to be worth a great deal more, as may be seen from this quotation taken from the columns of a scientific paper of the time: Tungsten $36 .00 per lb. Vanadium 10,780 .00 " Molybdenum 245.00 " Titanium 1,100.00 " Chromium 490.00 " THE PROCESS OF MAKING HIGH-SPEED STEELS 63 hardening metals is not infrequently above 20 per cent, it is seen that the cost of material alone is something quite different from what it is in the case of ordinary steels. Possible towered Cost. — In spite of the keen competition among makers, the price of high-speed steel has remained, practically where it was when first put upon the market, for the best grades not far from SO. 70 per pound, in small quantities. The so-called "new" steels sell for considerably more. This seems altogether out of proportion,' at first thought, even when the high cost of manufacture is considered. It is to be remembered, however, that this includes the as yet high cost of marketing; and must, of course, cover in part, also, the great expense of continued experimentation necessary to determine the most desirable composition and method of production. The cost of certain of the hardening constituents has increased considerably of late, owing to the large demand; but it may be expected that new sources of supply will be located and the methods of extracting the metals and ferro alloys will be so simplified that this cost will be materially reduced. It is becoming known, also, that the more rare of these agents are by no means essential to a good grade of high-speed steel; for other and less costly ones can be combined so as to give satisfactory results, especi- ally for that intermediate class of tools of which the highest duty is neither required nor desired. Apparently the marketing of the new steels is as yet one of the most costly items to the makers. As the place of the new tools becomes more and more definitely established the cost of bringing it to the attention of users will of course decrease, and the total cost to the consumer will without doubt be more nearly commensurate with what would be expected, and the economic value of the new steels be correspondingly increased. CHAPTER VI FORGING THE TOOLS. Stinted Use of High-Speed Tools. — The demonstrated utility and all- round superiority of high-speed steel for tools in most lines of metal working, and in other lines also, would lead to the inference that they are used to the largest possible extent. A recent study of machine shop conditions has shown conclusively that while they have taken a very large place in productive industry, the new tools are not used to any- thing like the extent they might be with profit. In comparatively few shops is high-speed steel largely used; in most, to a very moderate extent only; while in a great many it is quite unknown. Unsatisfactory Experiences. — Conservatism of course plays a large part in this condition of affairs, while unfortunate and misleading experiences seem to be responsible for the indifferent or negative atti- tude in many quarters. That there have been unfortunate and mis- leading experiences is for the most part due to the unintelligent manner in which the new problems of using high-speed tools is generally at- tacked. Persuaded in one way or another to buy some high-speed steel, the management of a shop more than likely turns the stock over to the regular tool makers, who are in this case almost sure to be unfamil- iar with its properties and the methods of treating it — and more than likely also, prejudiced against such new-fangled stuff. Under these circumstances it is not surprising that tool makers so often fail to profit to the largest extent by the directions furnished with the steel. Even when these necessarily brief and incomplete directions are followed as faithfully as possible, the inexperience of the smith makes his efforts more or less experimental, and the results may or may not be satisfactory. Purchasing Tools to Specifications. — The obvious thing to do, as pointed out in another place, is to buy tools made to specifications, for such introductory experiments. This also is the proper procedure in small shops using few special tools, after they have gone into regular use. Makers are quite ready to furnish tools of any pattern to specifi- cations designating precisely the material upon which they are to work and the machine and other conditions under which they are to operate. In this way there can be little question of the results being satisfactory. Making Simple Tools. — The making of the common forms of lathe and planer tools is so simple a thing however that it can be undertaken 64 FORGING THE TOOLS 65 in almost any shop having tool-dressing facilities — provided the smith is willing to forget, for the time being, some of the things he already knows concerning carbon steels, and also to learn a few things which may be quite surprising to him, particulary if he has no experience with high-speed steel. His knowledge of colors, as a guide to heating and tempering tools, for instance, will be no guide at all, and is likely only to mislead him. Whether it be in a large or a small plant, unless the work is in the hands of a trained expert, the first experiences in the making of high-speed tools should be with the simpler forms already indicated. These usually require little in the way of special appliances in order to give fairly satis- factory results. In their making, nevertheless, is involved the same special knowledge as to treatment which is necessary in the case of more complex tools. Use of Tool-Holder Stock. — Formerly it was a common practice to use high-speed lathe tools in connection with tool holders, merely breaking off from the bar a piece of the desired length, grinding it to the required point, and inserting it in the holder. This is still largely done where the work is light and the speeds not intended to be very high. The unan- nealed stock was once quite generally used for the purpose. This stock, however, while very hard, has not (in the case of most brands, at any rate) passed through a proper hardening process, having acquired its hardness while cooling under the stresses of the rolls or blows of the hammers. To insure an even temper and the absence of strains which tend to imperfections and therefore short service, it is necessary to anneal the tool pieces, and then to harden them properly. For this reason, as well as for greater ease in separating the desired piece from the bar, the annealed stock should be used, thus avoiding the annealing process in making the tool. Cutting Stock from Bars. — High-speed steel bars which have been so annealed can be easily nicked and broken off, unless of large section. In that case it takes an expert to do it. Breaking however is not advisable, for it is likely to cause a disarrangement of the structure of the steel near the fracture, sufficient to damage the tool at that place. Frequently fine cracks are started which later develop, and eventually spoil the tool. It is safer, and when the nose of the tool is to be forged any way, causes no additional labor, to heat and cut off hot. In general also it is more convenient to forge tools of this sort before cutting from the bar. Where many pieces are used, especially if little or no forging is required, whether of the same or different lengths, it is cheaper to saw them off to length by a power saw. It is unnecessary to do this one at a time, for if clamped tightly enough and sawed close to the support, a large bunch of " tool holder " stock, for example, can be sawed through as if it were a solid bar. As between a band and a circular saw, the former is preferred. 66 HIGH-SPEED STEEL Most such cutting can be avoided by purchasing tool-holder stock ready cut to desired lengths. Advantages of Annealed Stock. — In making tools requiring more or less machining there is an additional advantage in the use of the annealed stock, in that it is readily machined — almost if not quite as easily as carbon steel used for the same purposes. In some kinds of tools this is of considerable importance. Again, the annealed stock is stronger, that is to say, tougher; and tools made from it are therefore better able to resist stresses in the neck or shank than if made of the unannealed, and on that account are less liable to breakage at those points. The tre- mendous stresses and strains accom- panying the use of these tools makes it important to look to this matter. The early complaints against uneven- ness have almost wholly ceased since makers, mindful of their own inter- ests as well as of the interests of the users cf their steels, have made a practice of sending out annealed bars only, except upon special order. The Forge Fire. — For forging, any good fire, a common forge fire among the rest, will serve; though indeed here, as in other cases, the better results can be expected where the better appliances are used. The first essentials are to secure the required heat and to keep air currents away from the tool while heating. This is accomplished in part by keeping a deep and clean fire. Coke is better than the ordinary smith's soft coal, the latter having a tendency to burn out too rapidly. Very good results are occasionally obtained in this way, though they cannot be expected as a regular thing. If a forge fire must be used, a hood of fire brick should be laid up over the fire to prevent radiation and the circulation of air currents. This hood makes it some- what easier to conform to another prime essential in forging high-speed steel, namely, that the piece be brought up to the forging heat gradually so the heat penetrates uniformly to the very center. This is exceedingly important and will be mentioned again. Fig. 46. Good type of ,gas forge. Made by the American Gas Furnace Co., New York. FORGING THE TOOLS 67 Advantage of Good Equipment. — Although it is possible, as has just been said^ to obtain good results with very primitive appliances, it does not pay to try to get along without suitable apparatus if even a few high-speed tools are regularly produced. These tools are of no especial value above ordinary ones unless they are made uniform and exactly to the specifications requisite for the various special duties to which they are set; and it is the height of folly to spend good money for tools indifferently made. Especially where many tools are requried, suitably designed furnaces and other appliances are absolutely essential to the obtaining of satisfactory results. Convenience of Gas Furnace — Coke Fire. — For forging, a gas furnace is unquestionably the most convenient; and in the long run it is prob- ably as economical as any other. Some users maintain it is more so, in the matter of fuel cost, maintenance and tool output. However that may be, the coke furnace, when properly designed, is very satisfactory and efficient. The gas forge has much to recommend it — convenience, cleanliness, minimum attention to operation and maintenance, ease of regulation, and uniformity of results, among other things. Customarily of the oven type, it is provided, as is the case with gas-hardening furnaces also, with air under slight pressure, say one to two pounds, and suitable means for control- ling the flow of air and gas and for properly mixing them in order to insure perfect combustion and economy of gas consumption. A Good Coke Furnace. — The coke fur- nace is much used, both for forging and hardening. Where the amount of work done is comparatively small, the same furnace will answer for both purposes, as also will the gas forge. A good form, easily built and as easily operated and maintained, is illustrated herewith, Fig. 47. Essentially it consists of a sheet metal or cast jacket enclosing the fire-brick heating chamber. Sheet metal fuel hoppers are at each side, so placed that the supply of coke or anthracite (either may be used, if in small lumps) is continuous and is fed directly to the top of the fire bed. The result is a hollow fire of uniform temperature, the fuel being gradually heated in its descent, to the temperature of the deep fire bed. The latter, filling the chamber to the fore plate, or perhaps slightly above, rests upon a sectional grate of cast iron, preferably arranged so as to be rocked when shaking down Fig. 47. _ A simple coke furnace adapted to either forging or hardening. 68 HIGH-SPEED STEEL ashes. The ash pit is also a wind box, whence air under slight pressure from the blower is forced up through the fire. The temperature is regulated to a nicety by the damper in the exhaust and the cut-off in the air supply pipe, and that with almost no attention. When not in use, the fuel is conserved by entirely shutting off the drafts, the amount consumed then becoming negligible. As the gases from a coke fire are almost or wholly non-oxidizing, a tool is not likely to be injured when heated in this way. An important advantage possessed by this form of coke furnace is that the heat is almost wholly confined to the fire cham- ber, so that there is no waste of fuel, or discomfort for the operator. Gradual Heating Necessary. — The heating proceeds at a moderate rate, neither too rapidly nor too slowly. In the former case the heat does not penetrate uniformly to the center and in the forging the steel does not flow freely under the blows of the hammer, with the results hereafter pointed out. There is danger also that cracks will be formed because of the strains set up through the unequal expansion of exterior and interior.- If the heating goes on very slowly the heat soaks up into the neck or shank of the tool, and when hardening takes place, unless the tool is annealed before that operation, this part has lost much of its natural toughness — a thing to be avoided, as already pointed out. In the case of unannealecl stock, which is hard anyway, the slow heating is of less consequence. The fire therefore must be clean and well sup- ported by good fuel in the case of a coke furnace, and well regulated in that of the gas furnace. In all cases it must not be too keen, for then the outer parts of the tool are almost certain to become very hot before the interior reaches a forging heat, that is to say, at least a bright red. It is of course, impossible to know precisely the interior condition of a heated piece of steel in ordinary practical operations, so that the smith must be guided very largely by his experience and judgment as to the proper time during which a particular tool is to be heated. It is safe to assume, in general, that a piece having a section not greater than one inch, if properly protected, or if heated in a good gas or coke furnace as already described, will be ready for forging when the exterior has reached a bright yellow. Effect of Uneven Heating. — No hammering should be done under any circumstances while the portion of a tool that is being forged is under a good red heat. Neither should the interior be considerably hotter" than the exterior, as is likely to be the case when the tool is large and forged on a cold anvil. The disregard of these cautions is almost certain to result in defective tools. The consequence is shown in the accompanying Figures 48 and 49. When forged with the center much hotter than the outside, the former remaining expanded to a greater degree than the latter when the forging is finished, contracts on cooling, with the result that there are minute openings within, while the outside appears FORGING THE TOOLS 69 F perfectly sound (Fig. 48). If, on the other hand, the outside be flowing freely while the interior is still too cold to forge readily, the outer portion on coojing contracts over the hard inside and in consequence there are likely to be many fine cracks on the surface, as shown in Fig. 49. Not infrequently these defects are not evi- dent, and make themselves known later, during the grinding or machin- ing (in the case of tools requiring this), or more likely during the hard- ening. Sometimes the damage does not manifest itself until the tool is set Fig. 48. What is likely to take place when the interior of a tool being forged is much hotter than the exterior. X^ rr ' 1 1 ^Kx r Fig. 49. Result likely to occur when forging with exterior of tool much hotter than interior portions. to its work, possibly not until sometime after, when greatly to the surprise of the user it sud- denly fails without apparent MjjLjsm cause. I Forging Temperature. — It is } '"::.,,.._, """--\ better to do all forging at an orange or even a canary yellow than below that temperature, for it in large measure obviates the danger of imperfect working just pointed out. High-speed steel is difficult enough to forge anyway, considerably more so than ordi- nary steels, though less so than the self-hardening steels; and it is well to keep it as ductile as possible. The various makers of these steels generally give very brief directions as to the forging heats to be employed, in a number | of cases indicating that a good red is high enough. In general it may be said that while this is usually high enough, the reasons already stated are sufficient to make the higher forging heat desirable in practically all cases. Even in the case of inferior high-speed steel the forging heat must be high enough Fig. 50. Heel of tool being drawn down under steam hammer to give support to the nose almost di- rectly beneath the cutting edge. A power ham- mer allows rapid forging of large tools and is on that account very desirable. 70 HIGH-SPEED STEEL so that there is no metallic ring under the hammer, only a dull sound. The temperature range recommended is something like a hundred or a hundred and fifty degrees from and above 1,000 degrees C, or from about 1,850 to nearly 2,250 F. Precise temperatures are, in forging, not a matter of particular concern, the colors being sufficient guide if care be taken to keep the temperature of the interior and of the exterior approximately uniform and well above the minimum bright red already suggested. Cautions as to Hammering. — The proper heat having been obtained, the forging is done in the customary manner. Inasmuch however, as Fig. 51. Tool bent down across edge of the anvil ("turned up"). high-speed steel works somewhat harder than ordinary steels, it is neces- sary to exercise some discretion with respect to the force of the blows. Indiscriminate hammering is likely to prove ruinous. A large piece needs to have the blows heavy enough so that their force sinks into the interior instead of being absorbed at the surface alone. On the other hand, a light tool would be ruined by heavy blows unsuited to its size. It is important also that the forging be done rapidly, so as to be completed in one heat if possible. This helps to avoid the troubles just described. Tools requiring considerable working, as in the case of the Taylor standard lathe tools, for example, generally require at FORGING THE TOOLS 71 least two, and sometimes three heats, according to the tool and the number of helpers or the use or nonuse of a power hammer. The latter not only saves labor, but is likely, especially in the case of heavy tools, to add considerably to the excellence of the forging. Successive Steps in Forging. — Concerning the successive steps in the forming of a tool, and the special methods to be employed, it would Fig. 52. Forming a bent side tool. seem scarcely necessary to say much. The steps are practically the same as those in the making of a tool from ordinary steel, and the methods may be about the same also, except that it is well to take into consideration the fact that the tungsten steels forge with greater diffi- culty than carbon steels do, and that in bending and similar work it is desirable to resort to certain expedients for facilitating the work, as shown in the accompanying illustrations, Figures 51 to 56 inclu- 72 HIGH-SPEED STEEL sive, 1 some of which show the methods recommended by Mr. Taylor. The clamp attachment for the anvil, shown in Figure 51 is of partic- ular interest. Its application will be readily understood from the illustration. Fig. 53. Forming a side roughing tool. Close vs. Rough Forging — Gages. — It is well to shape the tool as closely to the required form and dimensions as possible without over refine- ment, in order to save grinding. Of course there is an economic limit to the closeness of the forging, for when this approaches refinement, it is cheaper to grind. It is desirable to use gages freely for testing the form and size of tools as the work progresses. In some cases forms' have been used in connection with the anvil (Fig. 55), in which the shapes are forged with precision bu<", without great expenditure of time. 1 Figures 50, 51, 54, 56, 57 and a number of others throughout this book, are taken from Mr. Taylor's Address. Figures 52, 55, and several others also, are used through the courtesy of the Gisholt Machine Company. FORGING THE TOOLS 73 Customarily a combination gage, giving all the required angles of a particular tool, will be sufficient. Mr. Taylor, in his report already mentioned, describes and illustrates along with others, a surface plate and cone gage, here shown in Fig. 56. The plate has a hole in one corner, into which fit the dowels of various cones giving the desired angles. The limits gage, intended to be used in the making of the Taylor standard lathe tools, also is illustrated, at Fig. 57. This indi- cates the extreme limits within which the forging must be done. The limits may vary in different shops according to the adequacy or in- adequacy of the grinding facilities. The cheaper the grinding cost, the less accurate of course may be the forging. Fig. 54. Successive stages of forged and ground tools. Courtesy of Machinery. A very simple and convenient gage, Fig. 58, which, however, does not give precisely the actual lip slope, consists of a small piece of sheet metal giving all the angles of a particular tool. A surface plate is almost essential in connection with this gage. Not quite so simple, but very convenient, is the gage shown in Fig. 59. This resembles somewhat the Taylor limits gage, but gives only the minor limit of the nose form. In addition it gives, as the Taylor gage also might be made to give, all the other angles required. Of course, a set of gages is required, one for each tool made in quantities sufficient to warrant the expense. 74 HIGH-SPEED STEEL Fig. 55. A set of forming blocks to be used in forging lathe tools, furnished by Gisholt Machine Company with their tool grinder and form chart. Fia. 56. Trying tool against cone gage to test proper angle for nose. FORGING THE TOOLS 75 Form gages should provide for an allowance of T \ to I inch which is to be ground off the working edge of the tool. Even when gages are deemed unnecessary, this allowance is to be made by the smith; other- Fig. 57. Limit gage for forging the 1-inch Taylor round-nose roughing tools. Fig. 58. Simple gage used for testing angles of English form of blunt-nose tool. Courtesy of Samuel Osborn & Co., Ltd. wise the amount ground off will not be sufficient to remove all the burnt metal, and the tool will in that case work at a low efficiency until it has had several grindings. Many have noticed that, in their 76 HIGH-SPEED STEEL own experience, tools seemed to improve with use, at least for some time after being set at work, particularly if the grinclings were light. That is, after each grinding the tool seemed to last longer than before. Of course this is exactly what would be expected to happen when the tool has been insufficiently ground the first time. Each grinding removes more of the somewhat burnt outside portion, until the unin- jured metal is reached and the tool works at its highest efficiency. Fig. 59. Simple gage for standard roughing tool. Need for True Tool Bases. — It must be remembered that the tools now under consideration, mostly lathe and similar forms, are subjected to tremendous strains, and that they must on that account be held very firmly in position. For this reason, it is necessary that the side upon which such a tool rests in the post or holder shall be smooth and true, so that it shall be firmly supported. After hammering true on the anvil it is well also to grind the base. Guarding Against Strains — Re-annealing. — If reasonable care has been exercised during the forging it is unlikely that strains will have been set up. It is well, nevertheless, to guard against the possibility of them by re-heating tools to a bright red, holding them at this heat for a short time, and then allowing them to cool slowly in air or in dry ashes. This is to be done before the hardening, and after the forging heat has gone down below a black. The partial annealing not only helps to remove possible strains, but softens tools so that they can be FORGING THE TOOLS 77 ground with ease to their approximate shapes, or machined without difficulty, should this operation be necessary. It likewise anneals the neck or shank of a tool when this has for any reason been allowed to reach a high heat during the forging of the cutting portion. Grinding off Excess Metal. — Excess of metal, beyond what is neces- sarily removed after hardening for reasons previously given, is best ground off immediately after forging or annealing as above, on a dry emery wheel. This may be done while the tool is still red hot, or at any subsequent time before hardening. Danger in Stamping High-Speed Tools. — Attention has been directed to the possible results attending nicking and breaking off tools from the bar stock. It is unwise also to make any nicks or marks on a high- speed steel tool at any place where stresses are applied. All stampings or other marks, such as are customarily made for identification, should be put in places where cracks can do no harm. A mark or nick in high-speed steel acts very much as does a diamond scratch in a sheet of glass, and will, in a part of the tool subject to strains, often eventu- ate in its ruin. Even when placed where no harm apparently can come, it is advisable that the marks be no deeper than necessary to serve their purpose. Forging vs. Machining Tools. — It may be well to observe here that no tool should be forged which can, without prohibitive expense, be machined from stock. This generalization practically limits forging to lathe and other tools of similar form, made from the solid stock. Tools like punches may seem easily forged; but their tendency to burst, even when carefully forged, is a sufficient reason for turning down from stock rather than forging them. CHAPTER VII. HARDENING — THE HIGH HEAT TREATMENT PRACTICALLY APPLIED. Uncertain Results — " Over-Refinement." — As in forging, so in harden- ing, very crude apparatus can be utilized, sometimes with satisfactory results. For the hardening of an occasional tool only, it might be ad- missible to use the protected forge fire already described. But there would be no certainty in the results. A tool might, or might not, come out right. The only safe course is to use a properly designed furnace. If any considerable number of tools are used, a suitable equipment is indispensable if it is really desired to make tools which will exhibit the powers and advantages of high-speed steel to the fullest extent. The derision of over-refined methods, the feeling that tools " good enough " can be produced by common, crude methods, has no point. Over- refinement is of course possible, and the manufacture of tools can be made unnecessarily expensive. But it must not be forgotten that " good-enough " in the case of high-speed steel tools means, if it means anything, that the tool is properly made and treated, so that it works at its best and does not become in the end a very expensive tool by failing or by spoiling a lot of work. For all work where endurance and accuracy count for anything, that is to say, where tools need to be ac- curately sized and to stay so for the maximum time, as well as to work during a maximum period, refined appliances and methods represent money profitably invested. Oil and Coke Furnaces. — The coke or anthracite furnace described and illustrated in the preceding chapter, Fig. 47, is well suited to harden- ing high-speed tools, as is that shown in Fig. 60, herewith. There are also many other suitable coke furnaces in use. When used for harden- ing heats, it is desirable that there be some arrangement for suspending the tools just above the fire bed, to keep them from contact with the fuel. A fire brick hearth or floor can easily be placed just above the fire bed, and this will be very convenient in doing some kinds of work. The oil furnace is, in general, not suited to the hardening of high- speed tools. It is difficult to regulate the temperature or to keep it high enough; and ordinarily there is a good deal of oxidation. On finished tools this is particularly objectionable. The oxidizing action is in some cases partly obviated by the use of a baffle plate or a muffle; 78 HARDENING — HIGH HEAT PRACTICALLY APPLIED 79 and may indeed be wholly overcome by designing the furnace so that the tools are heated within a muffle or a crucible which in turn is raised Fig. 60. A coke furnace used in hardening high-speed tools at the royal small-arms factory, Enfield Lock, England. Fig. 61. Brayshaw twin-chambered hardening furnace, for oil or gas fuel. The illustration shows the furnace equipped for burning oil. The upper chamber is heated by waste heat from the lower, and is used for preheating. to a white heat by the rotating flames in the fire chamber. The flames must not be directed against the crucible, either in such an oil furnace 80 HIGH-SPEED STEEL nor in a similarly designed gas furnace, else holes are likely to be melted into the pot. An English furnace, Fig. 61, in which the flame is directed downward, toward the floor of the fire chamber, is claimed Fig. 62. Rockwell oil-burning furnace, complete with tank and blower. A self-contained outfit especially adapted to isolated duty. to be quite satisfactory; and at least one American furnace, Fig. 62, is claimed to have overcome the difficulties and to be well suited to this use. Gas the Ideal Fuel. — There is some diversity of opinion as to just which kind of fuel is best for high-speed steel heating, some maintaining that coke is not only ideal, but the only fuel which allows absolute control of temperature. On the other hand, the experience of others shows that gas furnaces are now made which will accomplish practically all that any coke furnace will do. This type is unquestionably the most convenient; and while it is true that the first cost of gas seems high, when everything is considered, it really is little if any more costly than other satisfactory fuels. The objection that in the gas furnace, as well as in others mentioned, oxidation of the tool takes place, has some foundation. It is true that, as often operated, the heating chamber of a gas furnace will contain more or less unconsumed air, and that some oxidation takes place as soon as the tools reach a high tempera- ture, above a moderate red. Most of this, however, is unnecessary in HARDENING — HIGH HEAT PRACTICALLY APPLIED 81 a properly designed and intelligently operated furnace, for the supply of air and gas will be so regulated that all the air will be consumed. The oxidation complained of not infrequently occurs because air currents enter the fire chamber through doors carelessly left open. Anyway, there is likely to be less of this scaling caused in the furnace than in the subsequent exposure in air-cooling, or in carrying to the quenching bath. With proper care all except those tools requiring the finest finish and the utmost precision can be hardened satisfactorily by using gas furnaces for the heating. Gas Manufacturing Plant. — This type of furnace can be used even where a supply of gas is not available; for fuel gas manufacturing Fig. 63. " An apparatus for producing gas from naphtha at a low cost. Desirable where gas is not available, or where the cost is excessively high. American Gas Furnace Co. installation. plants in size suitable for supplying an equipment of gas furnaces are obtainable at a cost and with an economy of production which makes them desirable even where artificial gas may be had at the customary price. The cost of gas is, generally speaking, in this way reduced at least a half. In any event, the cost of the fuel is by no means the most important item in the making of high-speed tools; nor indeed is it of 82 HIGH-SPEED STEEL great consequence in computing the net results. A single expensive tool spoiled for want of suitable facilities for hardening it, will pay for enough gas to heat a great many other tools. And with inadequate equipment many a tool is spoiled, or imperfectly hardened so that it falls below its maximum efficiency. Producer gas, it should be stated, has not been found well adapted to the production of such high tem- peratures as those required in hardening high-speed tools. Oil or coal gas is recommended. Gas Furnace Design. — Excellent gas furnaces are obtainable at mod- erate cost, and it is not intended to discuss here their proper design further than to point out a few important considerations. A fur- nace should be of such form that the heating chamber can be, if re- quired, entirely enclosed, to pre- vent radiation and fluctuation in temperature by entering air cur- rents. The gas and air should be supplied to the fire chamber al- ready mixed in proper proportion for complete combustion, and so directed that the heat falling upon the tools is for the most part that radiated from the fire-brick walls of the chamber or oven. It is de- sirable therefore that the flame be given a reverberatory movement by suitably curved walls or muffle plates in the heating chamber, or a rotative motion by a tangential arrangement of the burners or nozzles, so that it will be directed past rather than toward the center, where it would impinge directly upon the tool. The air supply must be at a pressure of between one- half and two pounds per square inch, the air blast inducting the gas. Both air and gas supply must be under perfect control. These considerations hold in the case of forging and oil-tempering furnaces as well as with those used for hardening. Electric and Special Furnaces. — More convenient than any other type of furnace, and more easily regulated, is the electrically heated; and this is coming into considerable use, especially for small work, testing, Fig. 64. An excellent type of gas-fired oven fur- nace. The flame impinges upon the under side of the floor upon which the tools rest. HARDENING — HIGH HEAT PRACTICALLY APPLIED 83 and the like. Ordinarily the cost of electrical energy, in the operation of a large hardening plant, runs rather high. For special forms of tools, specially adapted furnaces are desirable. For long and slender tools, like taps, drill, reamers, and the like, which are Fig. 65. An electrically-heated furnace for hardening small and medium-sized pieces. American Mnclnniai' Fig. -66. A vertical gas furnace for heating slender tools suspended by their shanks. Courtesy of American Gas Furnace Co. best hardened suspended from the shank, a cylindrical or rectangular vertical furnace is much better than an oven furnace. A modification of this form is suitable also for hardening in an empty crucible, as is 84 HIGH-SPEED STEEL sometimes done. It resembles, in its general features, the oil crucible furnace already described and is identical with that shown at Fig. 83 in the chapter dealing with the barium process. Other special forms also can be used to advantage where enough work is done to warrant their installation. Such is a special die-hardening furnace, which is designed to harden only the face of a large die. Oil-tempering and other Fig. 67. Stewart cylindrical (gas-fired) crucible furnace, Chicago Flexible Shaft Co. furnaces for relieving hardness or strains also are essential to a w T ell equipped hardening plant. These will be described in another place. The type of furnace to be used will, as may have been inferred, de- pend a good deal upon the kind of tool to be hardened, so that it is desirable to equip a hardening room with two or three different forms to meet the varied requirements. There will also be other appliances such as those for quenching, for example. Minimum Hardening Equipment. — The minimum equipment to be considered will include a combined forging and hardening furnace, of any of the kinds already described; and an oil-quenching bath, or a stream of air under slight pressure. The apparatus for air-cooling may be of the crudest form — nothing more than a pipe of any desired size (not too small, say not under f inch), leading from the' air supply and provided with a suitable cut-off, or pressure-reducing valve, if compressed air is used. Occasionally tools can be hardened with no cooling appara- tus whatever, merely being laid in a cool place, preferably where there is a current of moving air. This however, is taking long chances on tools, for no certain results can be expected under such crude conditions. HARDENING — HIGH HEAT PRACTICALLY APPLIED 85 A Moderately Complete Outfit. — A fairly complete outfit consists of a forge, an oven, hardening furnace, an oil hardening bath or air-cooling table, and an oil-tempering furnace. Both the latter are described in the paragraphs indicating their use. A well equipped shop for forging and treating high-speed steel tools, however, would contain the follow- (a) A forge of suitable size for ordinary work. Its use has been already indicated. (b) A medium (or large, according to the work to be done) oven fur- nace for the hardening heats. The small oven furnace or forge would Stewart combination gas furnace, with lead bath. be used occasionally, no doubt, for small pieces. This furnace could be used also for what annealing would be necessary in most hardening plants. It would, along with the forge, serve for pre-heating in con- nection with the barium bath and crucible furnace, as well as with the customary methods of hardening. (c) A cylindrical or rectangular vertical furnace of the kind already described, for heating long, slender tools which are best suspended from one end while being heated. If necessary, for the sake of economy, 86 HIGH-SPEED STEEL this furnace could be easily adapted so as to be suitable, when pro- vided with a crucible for that purpose, for hardening in barium chloride, or for hardening in a crucible without a bath. (d) A lead bath is very useful where there is a wide range of work, but is not essential in high-speed tool hardening, especially if a barium furnace or a crucible furnace using no bath is adopted. A convenient and economical arrangement is a lead bath on the same base with a forge and oven furnace, as shown at Fig. 68. This economizes space and is very convenient. (e) A cylindrical crucible furnace used without a bath. This is less convenient than an oven furnace, but is used in some plants because it practically prevents oxidation of fine tools while being heated. It does not, however, prevent oxidation, to some extent, when the tool is exposed to the air before or during cooling; and for that reason, among others, it is less desirable than the barium bath furnace. The lead bath, barium bath, and empty crucible furnace all can be so made as to utilize the cylindrical furnace body by interchangeable crucibles. It is of course more convenient to have a furnace of each kind likely to be much used in doing the sort of work in hand. In that case, this crucible furnace will likely be omitted, unless it is the intention to heat tools by this method as a regular practice. (/) An oil tempering furnace for " drawing " the temper of such tools as require this to be done after hardening. This is described in a later chapter. (g) A quenching bath or air cooling device. A very simple affair for cooling with air has been already referred to, which is quite good enough for rough tools of the simpler sort. For careful work a hardening table is desirable, and one suitable for this purpose is described in connection with the methods of cooling, as likewise is an oil quenching bath. Need for a Temperature Gage. — A pyrometer for gaging the tempera- ture and checking against the operator's judgment frequently, is es- sential to continuous good results. The novice, especially in the manip- ulation of the new steels, needs the guidance of such an instrument; and the experienced operator himself cannot afford to get along without it, especially when working with the barium or other bath process. For the latter, a pyrometer of the thermopile or the resistance type is generally used; while with the direct heating processes either of these, or a radiation pyrometer of the Fery type, can be used. It is well to have the fire ends (where this type of pyrometer is used) interchange- able on the different furnaces, or with a separate set for each, any one of which may be switched into the circuit with the indicator or recorder, whichever may be preferred. Pyrometers which will give uninterrupted good service under the intense temperatures to which they are subjected in this kind of work, HARDENING — HIGH HEAT PRACTICALLY APPLIED 87 are not easily obtainable. The fire ends break after being used but a few times; enclosing porcelain tubes crack and crumble; or the thermo- couples deteriorate and cease to work properly. In any of these cases the indicator or recorder of course does not register correctly, and there- fore is of small use, if indeed it does not mislead. When the fire end is suspected, it is well to check it with another pyrometer of known ac- curacy, or with clay temperature determining cones, sometimes called sentinel pyrometers. These latter are very convenient also in the absence of a pyrometer, to determine high temperatures. They are cheap, accurate and are obtainable in large variety. Each cone is numbered for identification, and melts down or fuses when the pre- determined temperature has been reached which the particular cone was intended to indicate. The cones obviously are not available for determining the temperature of a molten substance, as in the case of the barium or lead bath. Supplemental Equipment. — If any forging is done in the hardening room, even though not regularly, there will be also an anvil and the tools usually accompanying the same. The anvil illustrated in the previous chapter, in connection with the forging of a Taylor standard tool, is very convenient. There should be provided also a suitable variety of tongs and other appliances for handling the tools, some of course of the conventional forms used for the purpose, and others especially adapted to the han- dling of tools requiring to be heated all over or which for other reasons cannot well be handled by ordinary tongs. The jaws of ordinary tongs, covering as they do a more or less considerable portion of the sur- face of the tool in hand, affect the temperature of the parts so covered and likewise prevent their coming into free contact with the oil or air in cooling. The excellence of the tool is thus impaired, often to a con- siderable extent. Many failures to secure results with high-speed steel, to say nothing of ordinary tools, are unquestionably due to so apparently small a thing as this. In some instances tongs with in-curving ends, properly formed to grasp the tools, are useful; in others the jaws may be studded with projecting hobs or prongs so that but a small amount of relatively cold metal touches the tool. Various desirable forms will doubtless suggest them- selves as the occasion arises for their use. Arrangement of Hardening Room. — The arrangement of the various furnaces and baths will depend much upon the number to be installed, and the limitations of the hardening plant — among other things, whether or not carbon steel tools, or even other objects, are to be hard- ened also. Assuming that a hardening plant for high-speed tools only is contemplated, and is equipped with the appliances enumerated above, the arrangement would be somewhat like that shown in Fig. 69. At 88 HIGH-SPEED STEEL the extreme end would be the forge, and opposite it the anvil; next the small hardening furnace, if there be one, and the medium or large oven furnace; and beyond them the cylindrical and the barium furnace. Opposite these is the best place for the air table and oil bath, or either, if but one is to be used. On the same side with the quenching appli- ances and ranged at one side of them, are the lead bath, if there be one, and the oil tempering furnace. It is seen that this arrangement econo- mizes space and practically centers the furnaces about the air table and quenching bath. The circular arrangement is avoided, though it is rather more convenient, because the space in which the operator works will doubtless be found quite hot enough without having focused upon V Fig. Layout of hardening room of capacity sufficient for hardening all the tools used in a large manufacturing plant. him the radiation of all the furnaces which happen to be in use at one time. If coke furnaces are used, the arrangement would be different only to the extent that these replace the gas furnaces here contemplated. There would perhaps be fewer of them, but each would occupy more space. Provision for Ventilating. — Each furnace and bath should be provided with a hood, preferably telescoping so as to permit lowering or raising as occasion may require, to carry away fumes, smoke, and excess heat. It is desirable that the hoods be connected to a common vent which is exhausted by a fan. This will not only keep the room free of fumes, but will add greatly to the comfort of the operator by creating a cooling draft. The fumes from the lead bath at the high temperatures to which it is necessarily raised, and from the barium bath also under certain conditions, are very irritating and must not be allowed in the room. It is well also to jacket the furnaces. HARDENING — HIGH HEAT PRACTICALLY APPLIED 89 90 HIGH-SPEED STEEL Fig. 71. " Each furnace and bath should be provided with a hood, properly connected to an exhaust.' Individual hoods, one for each furnace, are to be preferred. Fig. 72. Hardening room, Standard Tool Company, Cleveland. The furnaces are surrounded by a continuous sheet-metal jacket to prevent the distribution of heat into the room. Light is made even by the baffle shutters. HARDENING — HIGH HEAT PRACTICALLY APPLIED 91 Heating Simple Tools. — Heating high-speed tools for hardening is a very different thing from heating them for forging, not only with respect to the temperature, but to the variation in method also. The way in which a tool is heated and quenched, in hardening, depends very much upon its form and the use to which it is to be put. Lathe, planer, slotting, boring and the like tools can for the most part be readily ground to shape after hardening, and are on that account the simplest to treat. The heating may be done in any of the furnaces already designated as suitable for the purpose. It has been done successfully also in an ordinary smith's forge, though as already pointed out this method is not reliable and is undoubtedly responsible for many disappointments and failures. If no better means of heating are at hand, the forge fire should be covered with a hood, as already described in the chapter on forging, and the bricks well heated before any tools are placed in the fire. Gradual Heating Required. — When using a fire of this kind, or a coke furnace alone, it is well to place a number of tools toward the edges of the fire or upon the ample foreplate provided for that purpose in the case of the coke furnace, bringing each in turn nearer to the hottest part of the fire. This allows of slowly bringing the temperature up to a bright red, about 1,000 degrees C. (1,800 F.). When this heat has been reached the tool may then be rapidly brought to a dazzling white, anywhere above 1,200 degrees C. (2,200 F.), so the surface begins to flux and the corners and edges show signs of melting down. A few steels will harden properly somewhat below this temperature, and it is well to note and follow the direc- tions of the makers on this point. There need be no fear of over- heating, for as a rule no good high-speed steel is injured by any heat to which it can be subjected in any fire such as has been here described. Time and Extent of Heating. — The time required for bring- ing a tool from the red to the white heat will of course vary with the size of the tool and the intensity of the heat. Under gOOd Conditions it Should not " ba( : k than neC e S sary. The line AB indicates ap- need to take more than two ^5?^*^ to which ^ tool '' hould be minutes for a one-inch tool. It is important that the heat soak into the interior of the nose or working part so that it is uniformly hot throughout; and that while the whole 92 HIGH-SPEED STEEL of the nose is so heated, the heat shall not soak up into the neck of the tool. The white heat should not pass beyond the line AB, shown in Fig. 73. It is to be noted also that some makers of these steels recom- mend that the heating be gradual from the cold to the intensest white. This, however, does not seem to be really necessary, and it is usually more convenient to heat, in the way indicated above slowly to a red and rapidly afterward to a dazzling white. Using the Gas Furnace . — If a gas furnace is used to give the harden- ing heat, rather more care must be taken to keep the white heat in the nose of the tool. It is on that account desirable that tools of this kind be suspended through an opening in the top of the furnace, so the ex- tent of the heating can be controlled as closely as necessary by the distance to which they protrude into the heating chamber. It is de- sirable that the first or slow heat be given in a pre-heating furnace, the temperature of which is kept at or near a red heat, say about 700 or 800 degrees C. (1,300 to 1,500 F.). The tools are transferred to the high- heat furnace as rapidly as they can be handled conveniently. This serves the double purpose of preventing, in the case of large tools, the sudden lowering of the high temperature in the hardening furnace and the consequent need for regulating it again, and the blistering effect upon the surface of tools thrust cold into a furnace at white heat. Not only is the surface blistered under these circumstances, but the outside of the tool heats so rapidly that corners will be melted down and the tool will have the appearance of being ready for cooling when as a matter of fact the interior probably has not nearly reached the required tempera- ture. A tool hardened in this way very naturally would be defective. Grinding to Shape before Hardening. — In order to avoid excessive grinding of the hardened tool, it should be brought pretty closely to the required shape on a dry emery wheel after cooling down from the forging heat and before being subjected to the hardening heat. Some allowance must of course be made for the grinding subsequent to the hardening, to remove the burnt skin and restore the cutting edge. Precautions Necessary.— All tools with projecting edges, grained surfaces, sharp angles, or many clearances, are peculiarly susceptible to cracking during and after hardening unless this has been carefully and properly done. It is evident, therefore, that all possible precautions should be taken, by the use not only of care and intelligence in the treatment, but of adequate and approved special appliances whenever these have been shown by experience to help bring the best results. The need for certain special hardening furnaces indicated in a previous paragraph is made very evident when tools of the classes just mentioned are to be hardened. Suspending Slender Tools. — Such tools as long taps, reamers, drills, and the like, especially when slender, are very liable to warping and bending, unless heated (and cooled also) in a vertical position. They HARDENING — HIGH HEAT PRACTICALLY APPLIED 93 should therefore be suspended by their shanks during the heating. A vertical furnace such as has been already described, and illustrated in Fig. 66, is desirable for this purpose, though a coke furnace could have its top suitably arranged to allow of the same thing. The shanks of the tools project through holes in the cover of the furnace and they are held in place by tongs or holders provided for that purpose. The pre- heating can be carried on in any convenient way. Temperature Limits. — The temperature is not carried as high as in the case of tools which can be ground after hardening, and must always be short of the point where the cutting edges begin to melt. The limit of temperature is about 1,250 degrees C. (2,300 F.) except for heavy rough- ing cutters, when it is 50 to 100 Centigrade degrees higher, and may range downward to a hundred degrees below that point, or from a mel- low white or light straw to a bright lemon or very light orange color. Where the tools are of such a kind that the cutting edge can conveniently be re-ground after hardening, the heat may be carried up to that gener- ally given to forged tools, or a little above 1,300 degrees C. (2,400 F.). Tools for most lands of work are the better for this, if they will allow of the higher heat. The tools must not be allowed to touch the fuel nor be exposed to a flame after reaching a yellow heat, lest the cutting edges be injured. It must be remembered that, as in the hardening of all high-speed tools (except as pointed out in the chapter dealing with the barium process), the heating must proceed evenly throughout the tool or throughout that part which is to be hardened. Otherwise, strains are sure to be set up during the cooling, which are not relieved by the ordinary methods of tempering even, and which inevitably affect the endurance of the tool. All tools of intricate shape are pecu- liarly susceptible to cracking from such strains, the defects frequently appearing long after the tools have been set at work, if not immediately after the cooling. Watch the Pyrometer Indicator. — Success in hardening these tools depends very largely in getting just the right temperature in the heating. It is very necessary, therefore, to watch carefully the progress of the heating when the color begins to verge on a light yellow, so that the cutting edges shall not be damaged and a crust formed which would afterward need grinding off and thus affect the size of the tool. It is well to consult the pyrometer frequently at this point, for the varying conditions of light on different days and even in different parts of the same day are quite enough to affect the judgment of the operator. Milling Cutters and Like Tools. — Milling cutters and similar formed tools are heated in practically the same way as are tools of the kind just considered, except that the cylindrical furnace is not used. It is no better than the oven furnace for these tools, if as good. The cutting teeth or edges must however be kept from contact with fuel or furnace 94 HIGH-SPEED STEEL walls and floors, and it is well therefore to set such tools on end upon pieces of fire brick of appropriate size, or to suspend them from above by a suitable arrangement. Unless this precaution is taken the keenness of the edges is almost sure to be impaired, if indeed the hardening also be not affected. For the same reasons, more especially because the proper hardening is sure to be affected, in handling tools of this sort, tongs or other appliances must be used which will not touch the cutting edges. If such tools are of necessity heated in a forge fire, they should be, like drills and reamers, frequently turned; and it is well to do this however they may be heated. The cutting edges must not be allowed, when at a yellow heat or above, to rub against the fuel; and it is better that they do not even come into contact with it. This is one of the reasons why a forge fire is not well suited to the hardening of fine tools. Another reason is the oxidation which inevitably takes place to a greater or less extent under such crude conditions. Oxidation and its Prevention. — To protect such tools, heated under these conditions, from scaling and impairment of edges, a file-maker's paste, sometimes called a hardening paste, has been used by some. This however, while possibly serving the purpose desired to a consider- able extent, leaves the surface of the tool unclean, so that the hardening is not infrequently affected. The oxidation trouble is often very annoying, when it is not prevented, necessitating the re-grinding of tools after hardening and consequently also necessitating making them in the first place enough larger than the finished size to provide for this contingency. Except in the forge fire, oxidation need not occur to any considerable extent in any properly designed and intelligently operated furnace. Ordinarily most of the oxidation takes place after removal from the furnace and while the tool is exposed to the air. It is desirable therefore that tools be not carried, exposed to the air, any considerable distance for the cooling. This is imperative in the case of fine tools and those with sharp edges, unless they have been heated in the barium chloride bath. Lead Bath and Pack-Hardening. — It is to prevent oxidation entirely that the lead bath, the empty crucible muffle furnace, and like means, have been resorted to. These will be further considered in connection with the barium process, in a separate chapter. The " pack-hardening " of fine tools serves its purpose quite effectively, but is now little practiced be- cause the same results can be obtained by less troublesome means. Where adequate facilities for getting the same results in a quicker and more cer- tain way are wanting the method still serves a purpose. The usual practice is to enclose the cutters, if small, in a piece of wrought-iron pipe, packed closely with charcoal, fine coke, or other customary pack- ing, with the ends of the pipe sealed with clay. If much of this sort HARDENING — HIGH HEAT PRACTICALLY APPLIED 95 of work is to be dojie there should be a suitable pot, preferably of wrought iron. Cast iron will do, but it must be expected that the bottom will drop out occasionally, so intense is the heat required. Tools placed in the packing case should not touch one another, and, where this can be done conveniently, should be suspended by a common support before packing, to facilitate their subsequent removal and quenching (Fig. 74). The pot and contents, after sealing up, are placed in the white-hot fur- nace until the whole is at the uniform high heat necessary for hardening the particular kind of tools in hand. No rule can be laid down for the length of time required, since that will depend entirely upon the size of the tools and pot. The operator must be guided by experience — and the pyrometer. Some indications as to the condition of the tools may Charcoal or Coke The Engineering JUagasine Fig. 74. Pot for "pack hardening" tools, with method of suspending cutters for convenient handling and avoidance of contact in quenching. be obtained by the old expedient of withdrawing from time to time wires previously inserted in the pot for that purpose. The contents having reached the necessary temperature, the pot is withdrawn, and its contents removed and quenched as rapidly as possible. Causes of Scaling. — It may be of interest to mention the causes of scaling or oxidation. The explanation is very simple. At high heats iron and oxygen (which latter constitutes about a fifth of the atmosphere) have a keen chemical affinity for each other, and the oxygen of the air attacks the hot iron (or steel) with great avidity. The resultant of their chemical combination is a scale constituted of iron oxide, which is the same as common red iron rust except that the latter contains some water while the former does not. Scaling takes place also when steel or iron is left in contact with fuel through which air is passing or with which air is mixed. Hence the need for the cautions previously given with reference to such contact. Special Methods. — The need for exceeding care and the use of suitable appliances in the hardening of high speed steel tools, in order to insure proper and uniform hardening and to avoid deformation of their shape, 96 HIGH-SPEED STEEL has been already referred to several times. Nevertheless its importance warrants still further mention of the matter, especially in connection with the use of gas or other furnaces with floors or hearths. Tools of compact shape can, when so heated, be placed directly upon the floor of the heating chamber, or preferably upon pieces of fire brick; and are removed in the ordinary manner, the tongs of course being suited to the purpose. Long and slender tools, and others which are likely to be deformed, may be laid upon bricks or bars which have been carefully leveled, and are removed by the aid of pronged bars or other suitable implements. To expedite the heating of many small tools, say like inserted cutter blades, they are laid upon parallel bars which in turn rest upon the floor of the furnace. A flat bar or suitably pronged rod can be used for placing or removing a whole row of them at once. In all these cases the implement used for handling the pieces is to be well heated up before lifting them out, for reasons already assigned. Punches, Shear Blades, etc. — For hardening punches, punch dies, shear blades, forming dies, and a variety of other tools more or less like them in use, the heat is not brought as high, generally speaking, as for those classes of tools already considered. These tools preferably are ground closely to shape before being hardened. The temperature is in all cases kept below a clear white heat — say at a lemon color or near 1150 degrees C. (2100 F.). From this it may range downward, according to the brand of steel used and the size of the tool, to a very bright red, about 950 degrees C. (1750 F.). Small shear blades hardened at the higher tempera- ture named give excellent service without being tempered. Chisels and other tools subjected to repeated shocks are taken at the lower tem- perature mentioned. Summary of Hardening Temperatures. — For convenience of reference the temperatures required for hardening the various kinds of high- speed tools are here summarized. Turning, planing, shaping, slotting, boring, and the like tools for roughing and medium cuts: a full to a dazzling white, as high a temperature as can be given without actually melting the tools. Melting does not occur, in most high-speed steels, below 1400 degrees C. (2550 F.). Milling cutters and similar tools for heavy roughing: a good white, 1300 or 1350 degrees C. (2375 to 2450 F.). Milling cutters for moderately light and finishing cuts, forming cutters, screw machine tools, tools for fine finishing, and those which are to hold keen edges where the strain is not great, tools for cutting brass, and nearly all woodworking tools: a mellow white or light straw, or a little deeper, say from 1250 to 1200, or even 1150 degrees C.(2300, 2200, 2100 F.). Twist and flat drills, reamers, threading dies and taps, and other tools subject to severe torsional strains: slightly lower than that given above, or say a little below 1200 degrees C. and down to about 1175 (2200 HARDENING — HIGH HEAT PRACTICALLY APPLIED 97 to 2150 F.) or slightly below. This would give a light lemon color, verg- ing into straw. It will not greatly matter if these tools be heated quite as high as those in the class above, though in general rather better results will follow if this difference be observed. Shear blades, punches and punch dies, stamping and forming dies, pneumatic tools and others subjected to repeated jars or blows: 950 to 1150 degrees C. (1750 to 1900 F.) or from a bright cherry red to a light orange or lemon, according to the shape and use of the tool. Light punches and snap dies would be given the lower heats, as also would tools like file-cutting chisels. Permissible Temperature Variations. — It should be remembered in connection with the above summary that the hardening temperatures of high-speed steels vary more or less according to the composition, and that it is well to observe closely the instructions of the makers relative to this point, or better still to make careful determinations when any given steel is to be used, and thereafter to observe the limits found to be most satisfactory. In the nature of the case the above determinations are only general; but it is asserted with confidence that but little vari- ation will be found desirable in the case of any high-speed steel of the now accepted standard composition — if it is not premature to speak of a standard composition. Quenching Agents — Water. — For cooling high-speed tools either air or oil is used to good advantage. Cold water is, in general, to be avoided in cooling high-speed steels, whether of the so-called " new " varieties or not. Quenching in cold salt water is of course possible, and has in some cases been recommended by makers. Nevertheless hot high-speed steel and cold water are not a safe combination. A tool so cooled may not crack; and indeed may perhaps be repeatedly quenched in water — and again it may crack the first time, or strains may be set up which will cause cracks later. The uncertainty of the method, if nothing else, makes it a good thing to avoid except possibly in those cases where extreme hardness is requisite and the danger of cracking can be over- looked. Hot water, speaking in a general way, is less likely than cold to cause cracking, and has been used successfully for obtaining extreme hardness. It is best kept at a temperature rather above 70 degrees C. (160 F.). The novice will do well to let water alone as a quenching agent. Air Cooling. — In the early days of high-speed steel air was recom- mended by most makers, to the exclusion of oil. It is coming to be pretty generally agreed now that if oil does not give better results, as some maintain, it at least does give quite as good as air, and that it has some advantages not possessed by the latter. Inasmuch as most high- speed steels harden by mere exposure to the air, little apparatus is abso- lutely required, as has been already noted. Some rather good results 98 HIGH-SPEED STEEL have been obtained in this simple way. The hardness of these steels, however, depends a good deal upon the rapidity and the method of cool- ing, on which account mere exposure to the air does not bring out the qualities of the tools to anything like their highest degree. For many tools, therefore, this method is out of the question. To obtain uniformly good results the air should be cool and in motion. Preferably it is supplied in a continuous and rapid stream, large in volume rather than high in pressure. Compressed air is better than that from a blower. The pressure must, however, be reduced to two or three pounds only at the nozzle. Apparatus for Air Quenching. — For hardening an occasional tool, as has been already indicated, nothing further is required than a supply of air coming from a suitable nozzle of ample size. The tool is held in the blast and turned continuously until cold enough to handle, when it is laid aside in a dry place. Where many tools are to be hardened, even if only of the simplest kind, it is very desirable that there be a cooling table where the tools can be me- chanically held and turned while the air blast plays upon them. Such an arrangement is almost in- dispensable in the case of rotary cutters. A cooling table of simple Fig. 75. Table for air-hardening revolving cutters, as . - . . . used at the Royal Small-Arms Factory, Enfield design, Used in the British Koyal oc ' ngan • Small-Arms Factory at Enfield Lock, is shown in Fig. 75. It consists essentially of an iron-top table provided with a rotating plate and spindle between two movable nozzles from which the air blast issues The spindle and plate can be pro- vided with a clamp for holding lathe and similar tools also. In cooling milling cutters and the like, the nozzles are turned to one side of the center of the cutter so that the air will impinge upon the pro j ecting teeth in such a way that they will act as vanes, and the cutter be therefore rapidly rotated by the air current. All cutting edges are in this way cooled with absolute uniformity. An air box, resem- bling that illustrated at Fig. 76, is desirable for cooling lathe and similar tools. L Air Plpo Fig. 76. Air box for air cooling of high-speed tools. Convenient for hardening lathe tools and those of similar shape. HARDENING — HIGH HEAT PRACTICALLY APPLIED 99 6 6 A 6 Air bubbles 6 DeUdubU "Ira act toskei 4 4 4 4 4 «, Wlr* oet to catch tool! dropped Net U itre>beoed bj croto b*i» 6 d / N«t frame of Angle Iron 6 6 4 « 4 • ^ Ug for support^ net 4 t 6 O if o \ Peif« Mod mppl; pipe tod Ittcnli The convenience and simplicity of this method of hardening certainly recommend it. There are, however, certain disadvantages. The cost of air, for one thing, is considerable, and not comparable with that of maintaining an oil bath. The first cost of the latter is also the last cost except for the negligible item of renewal. In the air blast, furthermore, in spite of the rapidity of the cooling and the exercise of the greatest care, there will frequently be more or less oxidation; and this is not per- missible, in fine tools at any rate, affecting their precision as it does. Scaling is unimportant in the case of rough tools, since they are well ground anyway after hardening. Tools cooled in air are, in general, rather slightly softer than those cooled in oil. Apparatus for Oil Quenching. — For oil hardening the apparatus may be almost as simple as for air hardening. In small shops where but few tools are treated, nothing more is required than a medium sized tank full of oil. The shop doing a good deal of hardening, however, needs a bath of ample size equipped with some device for cooling and circulating the oil. An excellent form of such an ap- paratus is shown in Fig. 77. It is seen to consist of a sheet metal tank of suitable size, having a sup- ply pipe and laterals at the bottom through which air under slight pressure is introduced. The pipes have small holes in their upper sides from which the air bubbles up through the oil, at the same time cooling and circulating it. A net for catching tools accidentally dropped is desirable, as also is a net basket at one side, into which small tools may be thrown from time to time, for quenching, without further attention. Kind of Oil to Use. — Various oils have been recommended for quench- ing high-speed steel, including linseed, cotton seed, rape, fish, whale, lard, tallow, paraffme, and even kerosene. It does not matter particu- larly, so far as the effect upon tools is concerned, which is used, as long as it is thin and does not become gummy. Some have certain disad- Fig. 77. Excellent design for oil-hardening bath. 100 HIGH-SPEED STEEL vantages, though, which it is well to consider. Kerosene oil has given better satisfaction than anything else in some hardening plants. It does not flash, as might be expected, upon the hot tool coming in con- tact with the surface unless the quenching is very awkwardly done. If the tool is plunged quickly to a point below the heated portion, or entirely in the case of tools heated throughout, there will be no flashing. Disadvantages of Certain Oils. — Whale and fish oil are excellent agents, but have offensive odors. These can easily be suppressed, however, by the addition of about three per cent of heavy (tempering) oil. This at first floats upon the surface, but usually mixes with the lighter oil in time. The hardening is not affected by the heavy oil added, and this combination is about as satisfactory as any could be. Linseed oil is too gummy for general use. Lard oil becomes more or less rancid in time, but is excellent; and cotton seed oil has practically no objectionable features. The point is not so much what kind of oil is used; but that the supply be ample to absorb the heat rapidly from the tool. Where much hardening is done it is of course necessary, as already noted, to provide a means for stirring and cooling the oil. Cautions as to Quenching. — The quenching itself seems, and indeed is, a simple matter. There are, however, some points that should be care- fully observed, to get uniformly good results. First, the quenching must be done rapidly. Not only is the tool to be plunged into the oil with the least possible interval between this and the removal from the fur- nace, to avoid oxidation; but the plunging itself should be quickly done. Circular cutting tools, like milling cutters, are plunged with the axis vertical unless the thickness is considerably less than the diameter. In that case they are quenched like thin dies; that is, in an upright position. Most other tools can be plunged with the long axis vertical. After immersion the tool can of course be turned to any position that may be convenient. The vertical plunging obviates to the largest possible extent the warping and cracking to which intricate tools, and even those which are not intricate, if carelessly quenched, are subject. A thin, flat die with relatively large surface, for example, if quenched so that one face strikes the oil before the other, even if the intervening time be infinitesimal, almost invariably is warped and becomes useless. Special Case of Slender Tools. — In the case of drills, reamers, and the like, the heating of course has not extended the full length of the fluted part (unless, as rarely happens, the whole length is intended to do work), and the quenching does not go beyond the heated portion, say not beyond where it is still a good red. This can well be laid down as a general rule: a tool, except as already indicated, should be plunged to a point rather nearer the edge or end than that to which it has been heated, and worked up and down slightly while cooling. In this way a distinct HARDENING — HIGH HEAT PRACTICALLY APPLIED 101 line of demarkation between hardened and unhardened portions will be avoided, and the snapping of tools at this place prevented. Large or Intricate Tools. — If of any considerable size the tool must be kept moving in the bath so that all parts immersed will be washed by- cool oil, otherwise the oil in contact with the surface becomes so hot that hardening does not take place properly. This is especially true of tools of intricate shape, with many recesses, or containing small holes. In the last named case the tool should be so moved in the bath that oil will flow freely into and through the openings. If several tools are quenched simultaneously care should be taken that they do not touch one another, lest the places touching fail to come into free contact with the oil and consequently do not harden properly. Essentials of the Hardening Method. — The method of hardening here described involves essentially this: The tool is heated to the highest temperature it will bear without injury to the cutting edge, and even to the melting point if it can be afterwards well ground. It is then quickly cooled in an air blast or in an oil bath. This process is simpler than that patented by Taylor and White, is much more used, and is quite generally conceded to give results equally good with practically all standard high-speed steels. Essentials of the Taylor-White Process. — The Taylor- White process consists in the following steps : First, the high-heat treatment. The tool is heated to the highest temperature it will bear, as in the general process already described. It is then cooled rapidly down to the " breaking down " point, about 850 degrees C. (1550 F.), and after this cooled more or less slowly, as may be convenient. The method recommended by the patentees, for the preliminary cooling, is to plunge the tools from the high heat into a lead bath maintained at a constant temperature of 625 degrees C. (1150 F.), and to hold them there until they have had time to reach that tem- perature throughout their entire mass. Mr. Taylor says * it is a matter of no particular importance whether the^ tool be cooled rapidly or slowly below the " breaking down " point; and indicates that it may just as well be cooled in the air blast as not, and does quite well if merely laid aside to cool in the normal atmosphere. Second, the low-heat treatment. The tool is reheated to somewhere between 375 and 675, say to approximately 625 degrees C. (1150 F.), pref- erably in a lead bath large enough to maintain a uniform temperature. The tool is kept at this temperature for about five minutes, and is then cooled, whether rapidly or slowly being a matter of indifference. Taylor- White and Other Methods. — The only essential difference be- tween the Taylor- White and the customary process is seen to be in the 1 See Appendix B. 102 HIGH-SPEED STEEL second or low-heat treatment, which is omitted in ordinary practice. In another place mention is made that high-speed tools do not run at their best until a short time after being set at work, after being " warmed up," so to speak. The warming up is not figurative, but real. The tool soon attains a temperature approximating the minimum above given, that is, 375 degrees C, and therefore accomplishes while at work what is intended to be accomplished by the low-heat treatment. The " self- treatment " thus received by a tool does not normally give so high a temperature as that recommended by Mr. Taylor, unless run so rapidly that the cutting edge becomes red hot — which is not good practice, generally speaking. The cooling naturally occurs when the tool is stopped preparatory to taking the next cut. Apparently, therefore, the second or low-heat treatment is superfluous. It is maintained, nevertheless, that the self-treatment just referred to does not accom- plish to the same extent what the low-heat treatment does, the tempera- ture to which the tool is raised being rather too low under ordinary circumstances. However that may be, the second treatment is all but universally dispensed with, and so far as can be seen without dis- advantage. Special Modifications. — A modification of the Taylor- White high-heat part of the treatment is sometimes recommended by the makers of particular brands of high-speed steel. The tool, after being brought to the requisite high heat, is transferred to a hot bath of some kind, whether lead, fusible salts, or the like, where it is cooled to a dull red, equivalent to a temperature near 675 degrees C. (1250 F.), or 690 degrees C. (1280 F.), according to one successful maker of many tools. It is then removed from the bath and allowed to cool naturally, or it may be rapidly cooled in an air blast or by quenching in oil. Mr. Gledhill recommends a still further modification, cooling to the point mentioned above or slightly higher, in the air or in a blast, and then quenching in oil. As a matter of fact it would seem that the manner of cooling is relatively of small consequence, except that if it be rather rapid in the first stage the result will be a somewhat better tool. But the high heat is absolutely essential; and the higher the heat, the better the tool — subject, of course, to the limitations already pointed out. Electrical Hardening. — High-speed tools may be hardened electri- cally, though the process has not come into very general use. No definite information is at hand as to the excellence of the tools so treated, though the results are said to be satisfactory. Two methods have been prac- ticed to some extent. In the first method, illustrated in Fig. 78, the tool forms the positive electrode of an electric circuit in which it is placed by being clamped in a suitable clip or holder. The other electrode is constituted of the walls of a cast-iron tank containing a strong solution of potassium car- HARDENING — HIGH HEAT PRACTICALLY APPLIED 103 bonate. There are, of course, the necessary fuses, switches, and, current regulators. The current having been turned on, the tool is gently lowered, into the solution to the depth to which it is to be hardened, FLEXIBLE CABLE Fig. 78. Arrangement of apparatus for hardening tools electrically by use of potassium carbonate bath. and moved, up and down a little so as to avoid an abrupt transition from hardened to unhardened part. The tool on entering the bath completes Fig. 79. Arrangement of apparatus for hardening electrically by use of the electric arc. The shaded portion in B indicates the location of the carbon point during the heating. The cooling is by air blast or oil bath, as in the ordinary method. the electric circuit, and an intense heat is set up in the part immersed. When this is seen to be sufficiently heated, the current is switched off and the tool allowed to cool in the solution as though in an oil bath. 104 HIGH-SPEED STEEL In the second method (Fig. 79) the electric arc is utilized. The tool is placed on an insulating block and attached to the positive electrode. The other electrode is a stick of carbon clamped in a safety holder. The current being on, at a low voltage, the carbon is touched to the part of the tool to be hardened, and moved about as desired until the required heat has been attained, the voltage being gradually increased through a suitable rheostat. The tool is then cooled in the customary manner. This method evidently is suited only to local hardening, and not to the general run of tools. CHAPTER VIII. HARDENING — THE BARIUM CHLORIDE PROCESS. 1 Preventing Oxidation.— The scaling or oxidizing of fine tools has been already referred to as troublesome in certain cases; and ways have been Fig. 80. A cylindrical gas furnace fitted for use with the barium chloride process. Crucible filled with melted salts and ready for use. 1 While the hardening of high-speed steel by the barium process was originated in Europe, it seems not to have been made commercially practicable until it was taken up in this country by the agents of the Firth-Sterling Steel Company (Wheelock, Love- joy & Co. in New York and Boston, and E. S. Jackman & Co. in Chicago) and per- 105 106 HIGH-SPEED STEEL pointed out by which this annoyance, and the attendant expense of re- finishing, could be minimized with such appliances as might reasonably be expected to form part of a moderately well equipped hardening plant. Even before the days of high-speed steel it was felt that there should be some means of entirely obviating the nuisance; and many methods have been devised to that end, a number of them entirely successful except for one thing. The coke furnace, the well-regulated gas furnace, and possibly other ordinary furnaces give, as already remarked, very satisfactory results except as to tools requiring a fine finish; and the Fig. 8L Good example of intricate tools readily hardened by the barium process. electric furnace, and the crucible furnace in which the tool is heated in a white-hot crucible, which in turn is heated preferably by gas or oil, entirely prevent oxidation during the heating process if proper pre- cautions are taken. So also do the lead bath and the pack-hardening method. But none of these methods takes cognizance of the fact that even though a tool may come out of the furnace absolutely undamaged by oxidation, the moment it is removed and comes into contact with the air it is immediately attacked and oxidation takes place to a greater or less extent according as the exposure prior to cooling to the normal temperature is long or short. To overcome the difficulty entirely it would appear necessary to quench tools of the kind indicated without fected for the treatment of their Blue Chip steel. It is stated by the makers of a few high-speed steels that their steels will not harden properly by this process. This is anomalous, if true. However that may be, the process, developed. only yesterday, as it were, and as yet doubtless capable of great improvement, has been already adopted or is now being adopted by all the leading makers of fine high speed steel tools. HARDENING— THE BARIUM .CHLORIDE PROCESS 107 bringing them into the air at all. This it has been impossible to do with the appliances in general use, until the discovery and practical development of the barium chloride process. Advantages of the Barium Bath. — There are, of course, other reasons for the use of a bath in hardening fine tools. One of the most important is the need for absolute control of the temperature to within a very few degrees, and absolute uniformity in heating through every projection and into every recess, in the case of intricate tools, especially when small. While ordinary furnaces, such as have been recommended, are in general very reliable, and their temperatures under sufficiently close control for most purposes, there are nevertheless some fluctuations occasioned by variations in pressure of the gas supply or of the air pressure, and in the size of the door and other openings during the heating. These fluctuations are insufficient in the case of large tools, generally speaking, to be harmful, because of the comparatively long time required to bring such tools to the required heat, even when introduced into the heating chamber after being preheated. The small tool, because of its little mass and consequently the short time required for heating, is liable to be brought to a temperature sufficiently different from the one intended, to affect its quality to a considerable extent. Furthermore, in spite of frequent turnings and other precautions, some parts of tools complicated in shape will heat faster than others by their closer proximity to the incandescent walls of the heating chamber, or because of the direction of the currents which circulate within it. This is of course of less conse- quence in large tools than in small or fine ones with delicate projections or keen edges. Difficulties in Use of Lead Bath. — Another difficulty, the remedy for which, in the case of certain classes Of tools, has been already pointed out, is that of warping while being hardened. Slender tools (like drills, reamers, and those of similar shape) of a size sufficiently large to admit of being thus heated, are not subject to this difficulty when treated in a cylindrical furnace. But small tools of necessity require a different method. The distortion is entirely avoided when the heating is done in a suitable bath. Lead, because of its high specific gravity, is not so well adapted for this purpose as certain others. The tendency is for tools to float to the surface, and thus be irregularly heated, unless held down by some means. The lead bath, while it has been successfully used for hardening high-speed tools, is held at the extremely high tem- perature required, with some difficulty. It begins to vaporize at about 640 degrees C. (1190 F.), and when heated much above that point rapidly volatilizes, giving off offensive and irritatingly poisonous fumes. These can, however, be conducted away so as to do little harm, if provision be made, as already shown, for effectively exhausting from a properly designed hood. 108 HIGH-SPEED STEEL There are other disadvantages in using the lead bath at these high temperatures, and indeed some disadvantages at any temperature. At a white heat the lead oxidizes rapidly, and even when the surface of the bath is protected by a thick covering of powdered charcoal, more or less of this takes place, the scum rising and floating upon the surface of the lead. A much more troublesome thing is the sticking of the lead to the surface of tools, and the consequent uneven hardness that results from the parts so covered cooling at a rate slightly different from the parts of the surface to which no lead adheres. Efforts to prevent this trouble only seem to aggravate it or to develop new ones equally objectionable or worse. Likewise, impurities in the lead not infrequently damage the surface of the tool with which it comes into contact, especially at the white heat to which it is subjected in hardening high-speed steel. Holes and interstices sometimes remain filled with lead when the tool is with- drawn for cooling, and the result is worse even than when flakes of lead adhere to the surface. Difficulties Overcome by Barium Process. — All these, and other difficulties are overcome by the barium chloride bath process. The chloride does indeed give off fumes, unless precautions are taken to pre- vent; and a thin coating of it adheres to the tool when it is withdrawn from the bath. This latter, however, is just what is required in order to prevent oxidation while the tool is exposed to the air; and since the film is evenly distributed, there is no uneven hardening. It is possible also to maintain a more uniform temperature, since the melted barium chloride circulates freely, much more so than the heavier lead, so that the temperature throughout the bath does not vary sufficiently to be taken into account. Heating in a fluid is little or no quicker than in an open fire or in a good furnace. Evidently, however, if the bath itself is uniformly hot throughout, the heating of the tool must be absolutely even. Pro- jections cannot be melted down nor burnt before the interior has had time to reach the same heat as the outside, since it cannot get hotter than the bath, and that is kept uniform at the temperature required for the kind of tools in hand. The danger of blistering the surface or melting down the corners of a tool put into a white-hot furnace without sufficient preheating is entirely obviated. Even if the bath should for any reason be at a temperature high enough to damage a delicate tool thus suddenly subjected to an intense heat, the barium chloride has a melting temperature so high that a relatively cool object plunged into the fluid immediately causes a coating of it to solidify around the article. The coating of solid barium chloride then protects the enveloped article until its temperature rises sufficiently to melt it off. Furthermore, even though the actual heating of any given tool pre ceed little or no more rapidly, it is possible to gain a great deal of time HARDENING — THE BARIUM' CHLORIDE PROCESS 109 by the simultaneous heating of a considerable number of tools. A basketful of small or medium sized tools can thus be hardened just as well, just as certainly, and just as quickly as a single one. It might be supposed that tools so heated and quenched, that is in promiscuous contact with one another, might vary more or less in hardness. Such nevertheless is not the case, all coming out absolutely uniform. Fig. 82. A day's work. Good illustration of the various classes of tools to the hardening of which the barium process is especially adapted. The Furnace and Equipment. — The furnace for hardening by the barium chloride process may be of any convenient form which will admit the use of a suitable crucible. A vertical gas-fired furnace is preferred, one so designed (Fig. 83) that the flames are directed around rather than toward the crucible, enveloping it in a whirl of heat which is absorbed uniformly over its whole surface. If the flames impinge directly upon the crucible, there is danger of holes being melted into it when the heat is turned on, before the bath has become fluid and able to conduct the heat away rapidly enough. The crucible is almost necessarily of graphite. It should rest upon fire bricks so disposed upon the floor of the combustion chamber as not only to prevent the bottom falling out, but also to allow the flames to cir- culate freely about the under portions as well as over the sides. It should be so adjusted as to height that the top rises into the circular opening in the top plate of the furnace. The crevice may be luted up with fire clay or left open, the latter being the preferred method. In this case, 110 HIGH-SPEED STEEL however, the rim of the crucible must rise well into the furnace top, as shown in Fig. 84. The supply of air and gas, and the pressure, would be the same as for any furnace of similar type but used without the crucible and its SEALED Fig. 83. Cylindrical gas-fired furnace and crucible for use in hardening by the barium chloride process. Fig. 84. Fitting of crucible into Leaving the joint unsealed, as in erable method. furnace top. B, is the pref- VALVE I '5TOP PIN PI J SAFETf VALVE 2 DRUM <5'P/PE y+~ VALVE — ,-,- (WATER) %? I'PIPE Fig. 85. Reducer apparatus for use where air is delivered at a pressure higher than 1 \ or 2 pounds. Necessary when using compressed air in gas furnace. Courtesy of E. S. Jackman & Co. 1. Valve with an adjustable stop or gage on it. 2. Drum with petcock for draining off the water which appears when the air expands. 3. Safety pressure valve set for about K pounds to the square inch. 4. Valve for regulating the supply of air to the furnace. 5. Fitting to prevent possibility of high-pressure air backing up into gas supply pipe. HARDENING — THE BARIUM -CHLORIDE PROCESS 111 contained bath. Fig. 85 illustrates a desirable apparatus for reducing and regulating the air pressure. In operation valve 1 is opened to the stop, which should be set so as to admit just enough air to blow off gently safety valve 3 when valve 4 is wide open. The flame is con- trolled by valve 4, but valve 1 must always be open when the furnace is to be used. A good pyrometer should by all means be a part of the furnace equip- ment for hardening by this process, so that accurate determinations may frequently be made of the thermal condition of the bath. It is well also to have a fire-brick cover pivoted at one side so as to be easily and quickly swung over the top when desired. It is convenient to have a small opening in this cover, which in turn can be closed by placing a fire brick over it. The fire chamber of course should be vented, preferably into a stack. In this case, however, it is desirable so to fix the exhaust that there is a space between the opening from the furnace and the pipe or small hood, at which the flame from the furnace can be seen emerging. By watching this the mixing of gas and air can be regulated to a nicety. A very small flame indicates perfect com- bustion. When no flame is visible, the air supply should be reduced; and if there is a large flame, too much gas is supplied in proportion to the air. The flame should be just barely visible. Electrically Heated Furnaces. — Elec- trically heated furnaces are in successful use in some plants, and are stated to be economical when used for continu- ous runs. In this type of furnace the loss of crucibles is practically nothing, one lasting continuously for six months or more; whereas in a gas-fired furnace it lasts scarcely two weeks. It should be mentioned, however, that in certain hardening plants the electrical furnace is not favorably r garded because of the apparent tendency toward emphasizing the formation of a soft skin at the surface of tools hardened in the barium chloride bath. In such an electrically heated furnace the electrodes are of very soft Fig. 86. Electrically heated barium chloride furnace, as used by Ludwig, Loewe & Co., Berlin, who were among the first (if in- deed not the very first) to make use of the barium process for hardening high- speed tools. 112 HIGH-SPEED STEEL low-carbon iron, placed opposite each other inside the crucible, which latter is imbedded in a thick layer of asbestos or other non-conductor Fig. 87- Electric hardening furnace, switch panel and transformer. Courtesy of General Electric Co. jQax///ari//:/ecdrodd ~~J. 7d Transformer \ IZZZZ To Tramfor/ner Fig. Method of starting the electrical furnace. of heat. This layer in turn is surrounded by a thick wall of refractory material like fire clay and other insulating materials, and all are held together by a steel or iron jacket. In this way the heat is so completely HARDENING — THE BARIUM 'CHLORIDE PROCESS 113 retained within the furnace that at the end of a day's run the exterior is scarcely hot. There are of course the usual accessories, and a con- troller for varying the voltage and resistance and thereby the tempera- ture. A very low voltage, say from 5 to 60 or 70 volts, is employed in operating the furnace, the higher tensions being necessary only at first while the Baits are being melted. Thereafter the voltage does not usually exceed 25. Alternating current only can be used, the direct current setting up electrolysis whereby chlorine is liberated and attacks the tools immersed in the bath. The fumes of course also are increased. The fusion of the salts mixture is accomplished by moving a supple- mental electrode of carbon close to the appropriate iron electrode (Fig. 88) until the sparking has melted some of the salts, which latter then conduct the current. The resistance offered to the current heats and melts the adjacent crystals until the movable electrode has established a melted stream to the other iron electrode. After this the contents of the crucible fuse quite readily. During the melting of course a higher tension is required than afterward. With the controller the tempera- ture can be regulated to within something like 10 degrees C. In taking the temperature, and in hardening also, it is to be remembered that a rela- tively thin layer, half an inch or so, at the top of the bath is a little cooler than the rest, the difference varying some, but usually being near 10 to 20 degrees C. Methods of Operation. — When starting or renewing the bath, the crucible is filled with commercial barium chloride l mixed with a small proportion, say about two per cent, of sodium carbonate, commonly called soda ash. The two substances must be melted together, other- wise, especially if the crystals are used, dangerous explosions are likely to occur. The soda ash, in a way which does not seem to have as yet been investigated, prevents to a considerable extent the rising of chlorine fumes from the bath. These are offensive and very irritating when breathed, and also discolor the surfaces of tools with which they come into contact. The soda ash seems also to have some other effects as yet not well understood. It gradually becomes exhausted, and requires renewal from time to time. It must be remembered that in renewing it is dangerous to throw the ash into the melted barium chloride. The danger is minimized or averted if the ash is mixed with several times its 1 Barium is one of the small group of alkaline earth metals which includes also cal- cium (lime) and strontium. Magnesium likewise is sometimes included in the group. Barium never occurs free in Nature, its most common occurrence being in the natural compounds heavy spar and witherite, both of which have commercial uses. The metal itself has no present use in the arts, though intrinsically it is very interesting. It is moderately hard, of a yellowish color, fusible at about 240 degrees C, and burns in the air with great brilliancy. Commercial chloride of barium sells in quantity at about three cents per pound. It fuses at 890 degrees C. (1635 F.). The chemically pure does not melt so readily as the commercial. 114 HIGH-SPEED STEEL own bulk of the chloride before being added to the bath. Care must be taken that the proportion does not exceed that mentioned, otherwise the temperature of the bath is not so easily regulated. The boiling point of the bath seems to be lowered approximately in proportion to the excess of soda ash; and since it is very difficult, if indeed it is at all possi- sible, to raise the temperature above the boiling point, the tools cannot be heated high enough to be properly hardened. The bath should be renewed whenever it becomes sluggish. The melting should be slow at first. Once well started, however, the bath is rapidly brought to the required temperature, which varies more or less according to the class of tools to be hardened, as already shown. In general, the temperature will be somewhere near, and usually rather below, 1200 degrees C. (2200 F.), being raised above that point or lowered beyond it as required. The exceedingly high temperatures to which roughing tools are raised are unnecessary for the kind of tools to which the barium process is best adapted. Those high temperatures are sufficient to melt down cutting edges and affect the surface finish of tools, and one of the reasons for using the barium process is to avoid precisely this thing, or the possibility of it. Length of Immersion. — The bath being at. the proper temperature, small tools may be immersed and left until they are throughout the same temperature as the bath. The time required will necessarily vary accord- ing to the size and form of the tools, but in the case of small and regularly shaped tools it will range from a few seconds to a minute, or possibly two minutes. Larger tools of course require a longer time to become heated through; while those of a half inch section, or smaller, should be ready in less than a minute. The operator must learn to gage the time by actual experience. This is comparatively easy with the barium process, for, since the temperature of the bath is no higher than that to which the tool is to be raised, the latter is not damaged by remaining in the bath for some time longer than would be required merely to heat it through uniformly. It is well, nevertheless, not to leave tools in the bath for any considerable time longer than actually necessary. The Protecting Film — Quenching. — When withdrawn from the melted barium chloride, the tool is covered by a thin film, which serves to prevent the surface coming into contact with the air. It is this fea- ture perhaps more than any other one that gives to the barium chloride process its distinctive value. The tool can be quenched in oil without having at any time, from the moment the heating began, been exposed to oxidation. The coating of barium chloride protects the tool to a con- siderable extent also when the cooling takes place in an air blast, though it flakes off more or less and leaves spots exposed to the. action of the air. The better way is to quench in oil. Dies for drop hammers, and tools for other uses where they are subjected to concussions or severe HARDENING — THE BARIUM 'CHLORIDE PROCESS 115 jarring, are not quenched, as a general thing, but on removal from the barium chloride bath are allowed to cool slowly in the air. The film of barium chloride protects them from oxidation. Fia. 89. A tool withdrawn from the bath is covered by a thin film of barium chloride, which protects it from oxidation when exposed to the air. Preheating Large Tools. — All tools of any considerable size should be preheated before being placed in the bath, and in certain cases it is desirable that small ones also receive this treatment. Of course, where very great attention must be given to the absolute preservation of lines and surfaces, large tools also are plunged into the bath without the pre- heating. Where this is not absolutely necessary, some time is saved by the preheating, for the immersion of a large unwarmed tool of course chills the bath so that it is then necessary to restore the temperature to the required point. Avoidance of Temperature Fluctuation. — This happens also to a much less extent when a tool which has been preheated, is plunged into the bath, since the temperature of the tool is necessarily considerably below that of the bath. Evidently, then, it is desirable that the bath be ample enough to minimize fluctuations due to this cause. Small tools of course do not have any important influence in changing the temperature, and so far as this point is concerned may be put into it without preheating. It is very important that the temperature be carefully watched, and regulated as may be necessary. The experienced operator of course learns to judge very closely by its appearance and behavior whether or not all is right, but even he needs to check up his judgment against a reliable pyrometer from time to time. The influence of a passing shower even, changing the brightness of daylight as it does, is sufficient to make 116 HIGH-SPEED STEEL error easily possible in judging the temperature by the eye. The oper- ator with limited experience must of course be very largely guided by the indicator or record, as the case may be. Method of Preheating — Saving Time. — Heating tools in this way, pre- paratory to their being placed in the barium bath, effects a considerable saving in time when many are to be treated. Several are kept in the pre- heating furnace or bath and are given the higher heat treatment in turn. For preheating, any convenient furnace may be used, though the reliability and convenience of the gas oven furnace especially recom- mends it for this purpose. The lead bath also is very convenient. The heat is carried up to a low red, not above 600 degrees C. (1100 F.), and preferably somewhat below this. At this temperature no oxida- tion occurs, and it is perfectly safe to raise tools to this point in the gas furnace and then to carry them through the air to the barium bath. Obviously the preheating temperature must be maintained uniformly at the point mentioned, else some of the tools will get hot enough to scale more or less. Of course, when the barium process is used in hardening tools where this is of no consequence, the temperature in the preheating furnace can be as high as desired. Quenching Methods. — It has been mentioned already that the air blast disintegrates the film of barium chloride which adheres to the tool when withdrawn from the bath, and that, it is therefore better to quench in oil. For this purpose no special appliances are necessary. The oil bath already described in connection with ordinary hardening methods serves excellently. The net basket into which small tools can be thrown without further attention until they are removed, makes this tank particularly convenient for use with the barium process. Unimpaired Surfaces — Cleaning. — The oil, like the air blast, disin- tegrates the coating of barium chloride investing a tool when taken from Fig. 90. When the scales of barium chloride have been brushed off and the oil wiped away, the surface of the tool is as clean and bright as before heating. There is no impairment of edges, finish, or color. the heating bath. When the scales are brushed away and the oil wiped off, the surface is seen to be as smooth and every cutting edge as keen HARDENING — THE BARIUM -CHLORIDE PROCESS 117 and perfect as it was before treatment. Not only is the finish unim- paired, but even the color is almost exactly as bright and fresh as when the tool was first machined or ground. Even an expert could not tell merely by looking at it whether a tool had or had not been hardened. It has happened a good many times that purchasers have returned tools treated by this process, before trying them, thinking, from their appear- ance, that they had not been hardened. For cleaning off the scales a wire brush is desirable. If any of the barium chloride should stick to the surface or cling to corners and recesses, it can be readily softened by immersing the tool in boiling water for a short time. The scale then comes off without difficulty. Closely Sized Tools. — A number of things are possible * with the barium process which were only dreamed of before its development. High speed steel taps and threading dies, and tools used for similar purposes, have until recently left much to be desired. Almost invariably, when hardened by the customary methods, they lose size slightly or have a roughened surface which interferes with their perfect working. Further- more, the shrinkage is not at all uniform, in some instances varying several thousandths of an inch even in tools of the same diameter, by reason of imperfectly regulated heating conditions and the inherent imperfections of the usual methods when dealing with this class of high- speed tools. Difficulties Overcome. — The barium process entirely overcomes this difficulty. And not only can the size be maintained with almost abso- lute uniformity, but the tools can be hardened in such a way that they combine the greatest possible cutting powers together with a superior toughness of supporting stock, to prevent breakage under the high stresses to which they are thus subjected. This, with the circumstance that the size is not appreciably altered, that the finish is left perfect, and that the keenness of the cutting edges is unimpaired, especially adapts it to the hardening of many tools (those already mentioned, as well as many other kinds) to the making of which high-speed steel has not heretofore seemed well suited. Taps, threading dies, and other tools with overhanging teeth or cutting edges which, when properly hardened, are likely to break off or crumble, or when let down sufficiently to over- come this difficulty are too soft to last long, can have these teeth or cutters hardened to any desired extent while the body of the tool remains in the annealed condition. 1 It may be of interest to mention, in this connection, that the barium chloride bath is also excellent for hardening carbon steel tools. When so used, potassium chloride may be mixed with the barium chloride to form the bath, in the proportion of about two to three. The potassium chloride lowers the boiling point of the bath to near the temperature required for hardening ordinary steels, and thus reduces the danger of over-heating them. 118 HIGH-SPEED STEEL Method for Special Tools. — The method is exceedingly simple. All that is necessary is to plunge that part of a tool which is to be hardened, into the bath, preferably after the customary preheating, just long enough for the teeth or cutting edges to become thoroughly heated throughout to the required temperature, and then to withdraw it before the stock or body has had time to become heated enough to harden when cooled. The tool is then quenched in the usual manner. Cautions. — It is to be remembered that heating the exterior of a tool only, and then suddenly cooling it, as is required by this method, often sets up strains and causes flaws because the outside and inside portions have not, in cooling, had time to adjust themselves properly. A little care on this point will minimize the difficulty; and the subsequent " tem- pering" to which most tools of these classes are subjected, can be made to relieve any strains which may have been set up in the hardening. Hardening Dies, etc. — Dies, and other tools subjected to repeated blows or heavy pressures, can be hardened in a somewhat similar way, thus avoiding a trouble which it was not possible, before the develop- ment of the barium process, to circumvent — that of dies breaking or splitting open. A die may have its face hardened, as the cutters described above have their teeth hardened, by this part alone being placed in the bath, leaving about half of the body not brought to the high heat. This method is of course especially useful in the case of dies with rela- tively heavy bodies. Care must be taken, in this case, to move the die more or less, according to size, in such a way as to avoid a distinct line of demarkation between the hardened and the unhardened portions. Methods of Handling the Tools. — Not the least important thing in hardening by the barium chloride process is the handling of the tools. Tongs with spiked or serrated jaws have been mentioned as essential to the handling of high-speed tools during hardening. These are useful in handling some kinds of tools in connection with the present process. For the most part, however, it is desirable to suspend tools of any con- siderable size in the bath by wires, or in baskets. Grids or perforated metal plates have been used, though these are not so convenient nor so certain. When the wires are used it is of course necessary to take precautions to prevent tools slipping off. Once a tool drops to the bottom of the bath the operator will have his troubles recovering it. Tools with holes through them of course are easily handled by wire hooks. The baskets may be made of wire netting or of perforated sheet metal. Most operators prefer the wire. A rather surprising thing is that neither the suspension wires nor the much smaller wire (or the thin sheet metal) of which the immersion baskets are made, melt or burn away in the intense heat of the bath. The barium chloride seems to preserve the metal, for there is very little deterioration in the baskets. HARDENING — THE BARIUM CHLORIDE PROCESS 119 When a number of tools are thus heated in a basket, the latter is trans- ferred with its contents, just as if a single tool were being handled, to the oil bath and quenched at one plunge. As previously stated, the con- tact of the tools with one another does not at all interfere, as might be supposed, with their absolutely uniform hardening. Fig. 91. Methods of suspending tools in barium chloride bath. Difficulties. — As is the case with other methods of hardening, there are some difficulties connected with the operation of the barium process. The chlorine fumes, and the discoloration which tools occasionally & American Machinist, N. F. Fig. 92. Pair of tongs for handling dies. sustain from them, have been casually referred to. Such chlorination does no damage except possibly where tools are carelessly left exposed for a considerable time. The exhausting hood and the soda ash in the bath, with proper attention, will take care of this difficulty. Formation of "Bubbles" or Blisters.— A more troublesome difficulty is the formation of " bubbles " or " blisters," and of " pits " on the surface of tools under certain conditions. " Bubbles " seem to form under two well defined conditions, or at any rate appear to be of two distinct kinds. Sometimes a tool which has been preheated, or which was never thor- oughly cleaned after annealing, or other heat treatment, is covered with 120 HIGH-SPEED STEEL a thin film, or spotted with flakes, of iron oxide. This apparently melts at a temperature lower than that necessary to hardening, and in melting collects in droplets on the surface of the tool, to which they firmly adhere when cooling takes place. They usually are solid (though occasionally hollow), very hard, but do no particular harm, especially on rough tools, and may be ground off where grinding is permissible or possible — which it is not in case of many tools like taps, forming dies, and the like. The remedy, or rather the prevention, lies in greater care in preheating, the temperature for this operation being kept well below the red at which oxidation begins to take place; and in thoroughly cleaning those tools, even though not preheated, which may be coated with the iron oxide. " Bubbles " or blisters very like those described, but readily brushed off after cooling, also are of not infrequent occurrence. They are thought to be caused by molten droplets of iron oxide, floating in the bath, com- ing into contact with the surface of a tool and attaching themselves to it. Since they are so easily removed, and do no harm, no attention need be paid to them. " Pitting." — Much more troublesome than either of the " bubbles " mentioned, is the " pitting " which sometimes takes place. There sometimes appears, in this case, to be a melting away of the steel in spots, particularly along the edges or in projections from the body of the tool, leaving usually a slight hollow which sometimes is accompanied by a raised lump as if the metal had melted out and not floated away. More usually a " blister " is raised, beneath which a depression or " pit " is found. When such " pitting " takes place the tool is ruined and is fit only for throwing away or for working down to a smaller size. Investi- gation into the cause of this peculiar phenomenon has thrown but little light upon it. Inasmuch, however, as it seems rarely to occur, 1 except when the very highest temperatures, those not far below the melting point of high-speed steel, are used, it is suggested that it may be due to the presence in the steel of particles not perfectly homogeneous, which fuse at a temperature below the melting point of the homogeneous por- tions. No trouble of this sort is likely to be experienced if the tempera- ture of the bath is not allowed to rise above 1200 degrees C. (2200 F.) or thereabouts. This is amply high for hardening nearly all sorts of tools, except those just as well treated by other methods. 1 Something very similar, if not precisely like this, is of common occurrence when tools are heated by other methods, if extremely high heats are put on. CHAPTER IX. TEMPERING. Extreme Hardness not Essential. — Some allusion has been made to this peculiarity of high-speed steel, that the fitness of a tool does not necessarily depend upon extreme hardness, but upon the very different property of red-hardness, by virtue of which it resists the tendency to become soft or rub away under stress after having become considerably heated either through the heat generated in working or through being intentionally heated subsequent to the hardening process. As a matter of fact, extreme hardness is not infrequently a detriment to a high-speed tool, since it has, generally speaking, brittleness and internal strains for its concomitants. The latter are set up by reason of the unequal hardening of interior and exterior portions; and the damage they may cause has been repeatedly referred to. Only small and rather regularly formed tools are measurably free from them after hardening. "Letting Down." — It has been likewise shown that brittleness in tools whose working parts have overhang to such an extent that there is not sufficient backing to prevent crumbling of the cutting edges, is ruinous. A special method in connection with the barium chloride process has been recommended as in large part overcoming this difficulty. Even when this special method is used, however, and always (except as hereafter stated) in the case of tools hardened by other methods, it is desirable, and usually necessary, to " let down " this hardness and relieve the strains by a subsequent treatment, " tempering " or " draw- ing the temper." This treatment can be carried to any desired point and practically all strains relieved, and the toughness of the overhanging portions of tools restored to whatever degree required for maximum efficiency. The extent to which the temper should be drawn is deter- mined very largely by the nature and use of the tool, as it also is in carbon steel tools. About the only tools which are not benefited by tempering are those of the heavy lathe and similar types, with well supported blunt noses. These are customarily set at work after hardenings and grinding, without further treatment. The makers of some of the " new " high-speed steels say that tempering is unneces- sary for any tools made of those particular steels. Where tools are desired to be of extraordinary hardness this would be true whatever the steel used. 121 122 HIGH-SPEED STEEL Tempering not the " Low-Heat " Treatment. — The tempering here recommended does not correspond precisely to the low-heat treatment of the Taylor-White process, for the temperatures are considerably below the minimum limit of the range recommended by them in that connection. It corresponds almost exactly with the tempering of carbon steels, though the temperatures are, in general, somewhat higher, and perhaps maintained rather longer. Results with Crude Appliances. — As in forging and in hardening, so in tempering, results of a more or less uncertain sort can be obtained with very crude appliances, or perhaps with none at all. A forge fire has served on occasion, though it must be said, that such crude apparatus is not at all conducive to accuracy and uniformity in results. If the novice desires to do a little experimenting with no adequate appliances, this may be done with small risk by taking a piece of, say, tool holder steel, such as is used for light cutting, and heating it until it reaches a dark straw color, and then cooling in the oil bath or the air blast, as when hardening. A greater degree of softness is obtainable by allowing the piece to cool in air; and a still greater by raising the temperature until it reaches a green tinge, and then slowly cooling. Some tools are bettered by following this treatment by another, in which the tempera- ture is raised to a faint red, just perceptible in the dark. Exact Temperatures Necessary. — Important changes take place in high-speed steels within very narrow limits of temperature, not only within the hardening range, but within the tempering range also. And while the colors through which the surface of a piece of steel successively passes, while being heated, are indicative of the temperatures, it must be remembered that colors, especially when so slightly different as are most of those commonly named and used in tempering, are not only difficult to differentiate, even to keen and experienced eyes, but that the judgment concerning them is easily affected by variations in the light. The range of colors, as ordinarily used, extends from light straw to purple, or from 220 to about 275 degrees C. (425 to 530 F.). It is readily seen therefore, that exceeding care must be used in tempering to obtain just the required temperature and no higher. In order to do this suit- able appliances are requisite. Trying to temper high-speed steel in a forge fire, as a regular thing, is folly. A Simple Method. — An oven furnace can be used with a moderate degree of success, if it be capable of having its temperature regulated with extreme nicety and provided with a pyrometer for gaging the same. This is little cheaper, if any, than the oil process, though it is available where the amount of work is so small as not to warrant the installation of an oil furnace in addition to the general utility oven furnace. Oven Furnace and Sand Pan. — A better method, recommended by several makers of high-speed steel, and giving fairly accurate results when TEMPERING 123 the temperature is carefully regulated by frequent observations of the pyrometer, involves the use of a metal sand pan heated by a suitable gas or oil burner (or by other suitable means, for the matter of that), large and deep enough to contain an ample supply of clean and well dried sand. The tools are immersed in the sand and brought to the desired tem- perature without trouble, if the pyrome- ter is frequently read. The method is very good also where it is necessary to dispense with the use of the latter instru- ment, since the temper colors are readily observed. A satisfactory device, particularly use- ful where colors rather than the pyrom- eter are relied upon for determining the proper tempering heats, consists of a metal plate supported in any con- venient manner and heated by gas or otherwise, as shown in Fig. 93. The plate is covered by a sheet metal hood or oven. Method of Special Alloy Baths. — A method sometimes used, where the variety of tools is small and it is not desired to provide extensive equipment, consists in heating the tools in baths which melt at pre- determined temperatures. Care must of course be taken that the baths are kept as nearly as possible at the melting point. The difficulty of doing this makes the method somewhat uncertain, for a rise of a very few degrees is sufficient to give a different result. Any sort of melting pot sufficiently large for the tools to be tempered, an alloy which melts at the temperature required, and any suitable fire, preferably one that can be easily regulated, as a gas or oil burner, is all that is required. The tools are placed in the melted alloy and kept there from ten to twenty or more minutes, depending upon the size and form, until com- pletely and uniformly heated through. They are then allowed to cool in air, as is customary in all tempering. If the metal does not chill and set around the tool when it is first placed in the bath, the temperature is too high, and should be at once regulated. It is to be remembered that in order to keep the bath fluid it is necessary that the temperature be maintained slightly above the melting point. This should be considered when preparing the alloy suited to the kind of tools to be tempered, and the mixture so propor- tioned that its melting point shall be a few degrees below that which it Fig. 93. A tempering plate with sheet- metal hood or oven, as used in the Crossly Brothers Motor Works. 124 HIGH-SPEED STEEL is intended to maintain. The obvious disadvantage, further than that already indicated, that which very much limits the usefulness of this method, is the impossibility of varying accurately the temperature of the bath without keeping in stock a considerable number of alloys. Where more than one alloy is used it is of course necessary to preserve the identity of each in some way so as to avoid mistakes in the selection for a given use. Composition of Alloys. — It is exceedingly difficult to secure absolute precision in the measurement of high temperatures; and the deter- mination of the rather moderate ones used in tempering even is usually attended with more or less uncertainty. Even when the most perfect instruments are used, the personal equation of the observer enters into their handling and into the observations. The alloys and baths here given melt (or boil, in the case of the linseed oil) at points near enough the temperatures given for all practical purposes. The designation of colors for the several temperatures given is somewhat arbitrary, for color discrimination again is a matter of personal judgment and experi- ence, as well as of light conditions, as has been already pointed out. TABLE V. Parts. Temperature. Color of Surface of Steel at Tempera- tures Given. Lead. Tin. Cent. Fahr. 14 8 216 420 Very faint yellow. 15 8 221 430 Faint yellow. 16 8 229 440 Light straw. 17 8 232 450 Straw. 18.5 8 238 460 Full straw. 20 8 243 470 Dark straw. 24 8 249 480 Old gold. 28 8 254 490 Brown. 38 8 265 510 Brown, with purple spots. 60 8 277 530 Purple. 96 8 288 550 Deep purple. 200 8 293 560 Blue. — 302 575 Polish blue. Boiling lins eed oil 316 600 Dark blue. Melted lead 321 610 Gray blue. 332 630 Greenish blue. Tempering in Oil. — A well-equipped hardening plant will have a good oil-tempering furnace. This consists essentially of a tank of suitable size and form to contain an ample supply of oil, a furnace arrangement for maintaining the temperature at the requisite point, and a high heat thermometer, or preferably a pyrometer. Inasmuch as it is necessary to regulate the temperature within a very few. degrees, the furnace must be of a kind whose heat can be controlled to a nicety. A gas furnace is on that account preferred. There should be a basket of TEMPERING 125 perforated metal or of wire netting, in which tools may be placed for immersion. This will facilitate their removal, since all are removed with the basket, without any need for fishing them out separately. Method of Operation. — The basket and contained tools may be placed in the oil while the latter is still cold, though it is better first to bring the temperature of the bath up to some- thing like 200 degrees C. Since the main purpose of tempering is to relieve strains, it is well to avoid plunging Fig. 94. Good type of furnace for tempering in an oil bath. Oil could be used for fuel as readily Fig. 95. Cylindrical oil-tempering furnace with hood. tools into a bath much hotter than this, for to do so would heat the projecting and exterior portions so rapidly as to set up new strains which then have to be overcome. The gradual heating, allowing the heat to penetrate as it rises, softens the exterior portions enough to allow a readjustment of the molecules under stress, while at the same time it leaves the steel less hard and more tough according as the temperature is high or low within the tempering range. The temperature of the oil bath is raised to the point requisite to the tools in hand, and these are left immersed for some fifteen minutes or more, the time depending somewhat upon their size and shape. Large tools are to remain in the bath longer than is necessary for small ones. The latter are not harmed by remaining in the oil, in case it is desirable to temper large and small tools simultaneously. It is to be remem- 126 HIGH-SPEED STEEL berecl however, that only tools requiring the same degree of heat can be tempered at the same time. Large tools may be suspended in the bath in any convenient manner. When removed, the tools are allowed to cool down in air without further attention. Large cutters and dies often are left to cool off in the oil. Kind of Oil Required. — The oil used may be any rather heavy kind, that best suited being the so-called heavy or black cylinder oil produced from petroleum. Tallow is sometimes used. The former can be raised without trouble to as high a temperature as is necessary in tempering. Electrical Tempering. — Drawing the temper by the heating effect of the electric current is possible, though not much practiced commercially. The method is useful if the barium process is not available for hardening, in the case of such tools as milling cutters and others having a central opening and requiring the body to be tough while the cutting edges are left well hardened. A vise and mandrel suited to the work in hand are + Supply JIain Kheostat i Tool Under Treatment -f- Secondary + Primary j-^ A/WWWWW*- Fig. 96. Apparatus for tempering milling cutters electrically. connected to a circuit and resistance regulator, as in electrical hardening. The tool having been slipped over the mandrel, and fitting loosely upon it, the current is gradually turned on and the heat brought up to the required point. Evidently the central portion of the cutter is first heated, and also most heated unless the operation be continued for a considerable time; and consequently the outer portions retain their hard- ness to a much greater extent than the central, which latter are corre- spondingly toughened. It is possible to duplicate very accurately any previously adopted degree of temper in tools of the same form and size by merely using the same current regulation. Obviously, ' however, the process is rather limited in its applications because of the difficulty of accurately determining and regulating the temperature of the surfaces of tools of various shapes and sizes. The method is an improvement upon the simple and well-known expedient of heating tools of the kinds mentioned by the insertion of a hot rod, somewhat smaller than the TEMPERING 127 opening in the tool, and whirling the latter until the desired color appears upon the surface. Importance of Proper Tempering. — Certain makers of high-speed steels state that while tempering, after hardening, is desirable in the case of their particular steels, it is not essential. The statement is not suffi- ciently accurate. Tempering is not essential in the case of any high speed steel tools of certain classes, or of any class if only moderately effective work be acceptable. The proper tempering of carbon steel tools is even more particular than the hardening; and if this is not exactly true in the case of high speed steel tools, 1 tempering them to suit the requirements of the particular service for which they are designed still is highly important. The failure to get desirable results from the use of high-speed tools can almost invariably be traced to improper hardening or, more likely, improper tempering. It is exceedingly important therefore to have at hand suitable data, in order that the proper temper, as well as the proper hardening heat, may be given every tool. Such data have not heretofore been generally available, each user of high-speed tools having to depend almost entirely upon his own experience and observation. This is after all what must be depended upon in any toolmaking plant; and it is perfectly obvious that some system of recording accurately the treatment of particular tools is absolutely necessary in order to arrive at determinations in the many special cases that arise wherever many tools of various sorts are used. This matter will be further considered in a subsequent chapter. Tempering Temperatures. — The following data as to tempering tools are accurate for practically all good makes of high-speed steel, and cover in a general way the range of these tools. The temper of high-speed tools is, in general, drawn (when they are tempered at all) somewhat farther than is done with carbon steel tools, as may be seen below. Lathe roughing tools, and indeed all tools for heavy roughing, are left untempered. Large reamers, and drills with heavy stocks, 230 degrees C. (440 F.), equivalent to a light straw surface color. Ordinary drills, small reamers, and other tools of the sort necessarily having rather light stocks or bodies and subject to considerable torsional strains, 240 degrees C. (460 F.), a full straw color. Threading dies and taps, 260 degrees C. (490 F.), very dark straw or brown-yellow. Ordinary milling cutters, and the like, 210 degrees C. (400 F.), faint yellow. 1 Allusion has been already made to the claims advanced by the makers of some of the so-called "new" high-speed steels, that these require no letting down or tempering. 128 HIGH-SPEED STEEL Punches, stamping or cutting dies, and shear blades, 280 degrees C. (530 F.), purple. Chisels, snaps, and the like tools subjected to sudden shocks, 300 de- grees C. (570 F.), polish blue. Woodworking tools of nearly all sorts, 275 to 300 degrees C. (525 to 625 F.), light purple to greenish blue, according to shape and kind of wood to be cut. Brass working tools, 20 to 30 degrees C. lower than for iron or steel cutting tools of same kind. CHAPTER X. ANNEALING. Advantage in Using Annealed Stock. — It is now customary for makers to furnish high speed steel stock annealed, instead of as it comes from the hammers or rolls, as formerly, though in most cases the unannealed can be had if specially ordered. This is very greatly to the advantage of the user, since the bar stock thus becomes available in a variety of ways without preparatory treatment, it being, for example, a compara- tively easy matter to machine tools from stock, should this be desirable, as well as to forge them. The use of annealed steel lessens the need for unusual care in the matter of bringing tools out of the hardening treat- ment with shanks or necks soft enough to minimize the danger of break- age at those points. It is to be noted also that the long annealing at the mills gives to the steel a uniformity and freedom from strains which otherwise would considerably limit the utility of fools made from it. Most high-speed steels, can, by proper annealing, be made nearly or quite as soft as ordinary tool steels. The method by which this is accomplished at the mills has been briefly described in a former chapter. Annealing Furnaces. — Whether because only the unannealed stock is at hand, or because during the forging or other processes through /Pyrometer Stem The Engineering Magazine Fig. 97. Annealing furnace at works of the Brown & Sharpe Manufacturing Co, which it may have passed during the making and prior to the final hardening, a tool may have become more or less hard, very likely un- evenly, it not infrequently becomes desirable or necessary to anneal 129 130 HIGH-SPEED STEEL pieces in the toolmaking plant. If much of this is to be done a suitable annealing furnace of the necessary capacity of course is requisite. A gas fired oven furnace of ample size is most often used, though the slightly less convenient coke or anthracite fired furnace also is in all respects suitable and satisfactory. An important requirement in design is the ability to sustain a continuous high heat for a long time, if necessary. For ordinary annealing such as may be expected to be done in an ordinary toolmaking plant, even this is not essential, for the heats need not be of a duration comparable to that given at the mills. For occasional work the ordinary furnaces used in hardening are sufficient. Uniformity Essential. — A prime essential in annealing is uniformity in the result — a given piece must be annealed to the same extent through- out. Obviously, therefore, the heating must be uniform, and it is neces- sary that the furnace be such that the heat shall be evenly distributed throughout the fire or heating chamber. The coke furnace is espe- cially good in this respect, though gas furnaces are now designed with muffle or baffle plates, or double floors, so as to accomplish the same result. Methods of Rapid Annealing. — Quick annealing, especially with but indifferent facilities at hand, is not to be encouraged; for though fairly good results may sometimes be thus obtained, it must be remembered that proper annealing is a process requiring much care and good judgment. If necessary to anneal a piece without any of the desirable appliances, this may be clone with more or less success by heating slowly in an open fire or furnace to a blood red, holding there for some little time, and then allowing the heat to die down very gradually, the tool remaining in the fire or furnace until entirly cold. The slower the temperature is raised and then lowered, the better the annealing. This holds true with all methods in general use. If a smith's fire be used, care must be taken to make it deep and to cover it up well after the desired heat is reached, so it will die down very slowly. It is better to build a hood over the fire, resembling that mentioned in an earlier chapter, and to fill this up with coke before the fire is allowed to go down. If a gas or coke furnace be used, air must be carefully excluded during the cooling, to prevent excessive oxidation. The heating will occupy anywhere from two hours to ten, according to the size and condition of the piece, and the degree of softness required. The heat must thoroughly and uniformly penetrate the entire piece. Instead of cooling down in the fire or furnace, the tool being annealed may be withdrawn and buried in sand, ash, lime, or asbestos of generous quantity and previously well heated up, and allowed to cool there. A quicker method is sometimes employed, which, however, is not very certain, and but partially accomplishes the object. The steel is heated to a dull black red and plunged into hot water. The temperature of the ANNEALING 131 Fig. piece must not exceed that indicated, while that of the water must be little short of the boiling point. Protection from Oxidation. — If possible the steel should be enclosed in a muffle, even if nothing better be at hand than a piece of gas pipe, and packed closely in green coal dust, coke dust, powdered charcoal, or the like substance, to generate a non-oxidizing gas which shall envelop the pieces under treatment. Asbestos, ashes, or sand also serve pretty well, though usually the surfaces of the annealed tools are less perfect. It is much better to have a suitably designed muffle, preferably of cast iron and lined with fire brick, for this use. A rectangular box pro- vided with a flanged lid (Fig. 98) is excellent for the purpose. The flange should fit very loosely into a grooved rim projecting from the top of the box. The groove is filled with sand, fine ash, or fire clay, to exclude air. The clay answers this purpose rather better than the sand. It is well to have one or more small holes through the cover for the escape of gases; otherwise the clay filling may be blown off in places. It is very necessary to take precautions for the exclusion of hot air (and cold also, for the matter of that), because sur- face oxidation of the steel takes place, in that case, frequently to such an extent as to impair the uniformity of the subsequent hardening. Time and Temperature Required. — The box and contents are then placed in the moderately hot furnace and slowly brought to something more than a dark red, anywhere above the point at which softening is completed, and below that at which recalescence begins, or above 700 de- grees C. (1300 F.) and up to near 800 degrees C. (1460 F.), somewhat beyond which lies the critical point in heating. 1 This temperature is maintained from one to four hours, according to the size of the box and of the pieces being treated. The point is to be sure that every piece is completely and uniformly heated through. Soaking, that is to say, long continued heating, is likely to be injurious, and should be avoided. It tends to make the structure of the steel coarse, as it does in the case of ordinary steel, increasing at the same time the liability to cracking during the subsequent hardening. It is of course possible to arrive, by experimentation, at a definite time for which given bars or pieces must be heated in order to anneal perfectly and without damage to the structure; but in miscellaneous work the conditions vary so much that no specific rule can be laid down. Pot for annealing. 1 See the chapter on Nature and Properties. 132 HIGH-SPEED STEEL Only experience can be relied upon as a guide, though the limits above mentioned will indicate the approximate range of time within which the heating is to be done. It is better, at first, to under-heat rather than over-heat (that is, continue the treatment for a shorter rather than for a longer time than seems probably necessary), since the anneal- ing can be repeated in case it has not been carried far enough. Maximum Temperature. — More important still is the maximum temperature maintained. If raised considerably above that already indicated (800 degrees C. or 1460 F.), the purpose of the treatment is in part defeated because near this temperature the changes begin to take place which are necessary to hardening; and the steel comes out imper- fectly annealed, if indeed not quite hard at times, especially if the cooling proceeds less slowly than it should. The temperature may some- times, under conditions insuring extremely slow cooling, range a hundred or more centigrade degrees higher. Frequent readings of the pyrometer are very desirable. Cooling. — The heating having proceeded until the steel has been completely and evenly hot throughout, the heat is turned off and the furnace allowed to cool down very gradually — the slower the better — and the contents removed only when quite cool. This requires, under the best conditions, several hours. Twelve hours is a short enough time; and two or three times as long is much better. Cold air must, during this time as well as during the rest of the treatment, be carefully excluded from the furnace. Prevention of Discoloration. — The surface of high-speed steel so an- nealed comes out very well, but usually somewhat discolored. It is sometimes desirable to leave the brightness of the surface unimpaired. This is accomplished to a moderate extent by placing in the bottom of the annealing box a handful of resin and a second handful sifted over the top of its contents. A thin black film covers the pieces when taken out. This objection is avoided by keeping the contents of the annealing box constantly surrounded by an envelope of gas, air being absolutely ex- cluded during the whole time. To do this there must be a continuous supply of gas passing into the box, which latter is provided at one end with a small iron pipe screwed into a suitably threaded hole at one end and long enough to extend well outside the furnace. At the end of the annealing box opposite the pipe is a small orifice. The box having been filled and closed in the regular way, the pipe is connected with a gas supply. The jet issuing from the small orifice, after the- air has been driven out and the box filled with gas, is lighted and the box placed in the furnace to be heated in the customary manner. The gas, very little of which is consumed, takes the place of air in the interstices around the steel to be annealed, before the heating begins, and completely fills them during the entire treatment, so that there is no possibility of even ANNEALING 133 enough oxidation to color the surface of the pieces. The method is excellent where it is necessary to anneal finished tools. Electrical Annealing. — Experiments have been carried on looking to electrical annealing and to bright annealing by immersion in a bath of fusible metallic salts, somewhat after the manner of the barium chloride process for hardening. Moderately successful results have in some cases been obtained; but the methods are not as yet sufficiently developed for commercial use. The two methods have also been combined, with apparently good results, the salts bath being heated by the passage through it of an electric current. CHAPTER XI. GRINDING. Importance of Proper Grinding. — Hardening has been very generally emphasized as the one most important operation in the manufacture of high speed steel tools — the steel of suitable quality of course being granted. In one sense this doubtless is true, for a well-hardened tool will do moderately good work even when the remaining operations are badly done, or, in the case of some of them, quite omitted. It is equally true, nevertheless, that a tool giving maximum service must have passed successfully through all the operations necessary to its manufacture. These operations, from the bar stock to the finished tool, may be given as forging or otherwise forming, hardening, tempering, and grinding. Certain of these of course are simultaneous in some cases, as when a milling cutter is sharpened before hardening, at the same time it is formed. Of these operations, grinding is fully as important as any other, the same care and precision being required for the development of the highest possibilities in a given tool. As elsewhere in the manufacture of high-speed tools, incompetent or careless workmen, or inefficient methods, may easily spoil or render defective an expensive tool. It seems very certain that a large proportion of such tools, as made and used in ordinary practice, are more or less injured by improper grinding. The operation is in many places regarded indifferently, and in conse- quence there is a serious loss in efficiency. Undoubtedly a good many discouraging trials of the new steel turn out badly more on account of improper grinding than for any other reason. Of course not every tool plant can be equipped with the most approved machinery, and use the very best methods in grinding, any more than in the hardening and tempering of these tools. But it is to be emphasized that the best results can be expected only when efficiency characterizes every detail in their manufacture as well as their use. Kind of Stone to Use. — Much has been said as to the kind of stone best suited to the grinding of high speed steel tools. A good many makers of the steels recommend emery, or other composition stones; others prefer sandstone, and still others indicate that either may be used under suitable conditions. The fact of the matter seems to be that any of these several kinds of grinders can be used with satisfactory results — if the stone used be selected with reference to the work to be 134 GRINDING 135 done and the proper conditions be maintained. It does not follow that all are equally efficient. The thing to be desired in this respect is that the wheel shall grind as rapidly as possible, leave a sufficiently smooth finish, and yet not overheat the tool. A coarse and hard wheel ordinarily will grind faster than a fine and excessively soft one, say like a sandstone. In the case of artificially bonded wheels other than sandstone, a coarse, soft stone will cut faster than a fine hard one, the softness allowing the grain to break down readily, and thus continually to present new cutting grains to the surface being abraded, and the coarse- ness allowing large, sharp points to engage the work. For this reason, and also because it can be speeded up much faster, an emery wheel can be made to grind high-speed steel more rapidly than a sandstone. The ease with which artificial stones can be modified in grit and bond to meet the various requirements, gives them another advantage. While it is probably true that most of the damage suffered during grinding high-speed tools occurs when emery wheels are used, it by no means follows that the fault necessarily lies in the use of a wheel of this kind. As a matter of fact, such troubles usually result from the unintelligent Fig. 99. Chips from normal grinding magnified. ■* Courtesy of the Norton Company. use of the wheels. The inexperienced or inattentive operator is likely to forge.t that the emery wheel is running at a speed very much greater than the sandstone possibly can run, and that in consequence the tool pressed against its face is heated up much more quickly. The result is likely to be a ruined tool. " Glazing " and " Loading " of Stones.— A difficulty frequently met with heretofore in the use of these wheels, in grinding high-speed steel, has been their tendency to " glaze " and " load," that is, for the cutting 136 HIGH-SPEED STEEL grains at the grinding surface to become worn down smooth and dead even, and the pores or interstices in the bonding material to fill up with the abraded particles of metal. These two conditions sometimes occur separately, and at others together. When they do occur, grinding ceases in proportion as the surface of the wheel is more or less glazed and loaded. To grind, that is, to " cut," the grit at the surface of the wheel must be sharp, not worn down smooth; and the interstices be- tween the grains must not be filled up. A grinder is a sort of cutting tool. The cutters are the infinite number of sharp grains or grit whose angles are exposed at the surface of the wheel. These act individually very much like the corner of a broken file in scratching a metallic surface. They get behind slight inequalities in the surface being ground, or force themselves into the metal, and in either case push off a thread-like fila- ment (Fig. 99) whose size depends upon the size of the grit and the pressure applied. Evidently there can be no grinding under the conditions just stated, that is, where the grit is worn down smooth or the interstices of the grinding surface filled up to such an extent that the grit does not protrude. In hand grinding the tendency is, when one or both these conditions prevail, to press the tool the more firmly against the wheel. Since little or no work can be done with the grinding surface in this state, the additional pressure serves only to increase enormously the friction, and generally to ruin the tool by the sudden heating of its sur- face or cutting edge. Ruin of Tools in Grinding. — It may seem singular that a high-speed tool should be spoiled by any temperature to which it might be raised in grinding, for that is of course not comparable to the temperatures used in the forging or hardening. The damage comes about in large part through the " drawing " of the " temper," and in part through " checking " or the formation of surface cracks. Softening of high- speed steel, as previously shown, begins at a temperature approximating 550 degrees C. (1100 F.) and is completed near 700 degrees C. or 1300 F. The lower temperatures in this range are easily possible in careless grind- ing; and indeed, the higher, corresponding to a low red, has been observed at the point of a tool flooded with water. The more frequent injury probably is due to the manner in which the frictional heating occurs — the sudden rise of temperature in the thin outer skin of the tool face ap- plied to the surface of the stone, and consequently its rapid expansion without reference to the unheated portion back of and adjacent to it. The result is that numerous checks or cracks are formed, more or less deep according to the pressure, speed and duration. Under these cir- cumstances, if the stone be used wet the trouble is likely to be greatly aggravated; for the cold water coming into contact with the heated sur- face is almost certain to cause a multitude of checks over the whole surface affected. GRINDING 137 Wheels Suited to the Work. — A large part of the trouble, if not the whole of it, lies in the use of wheels not suited to the work in hand. It is well enough understood that the use of one form and size of lathe tool, standardized as much as it may be, for all sorts of jobs and on all kinds of material, is not only uneconomical, but exceedingly foolish. Various jobs require particular tools, such as are specially adapted to the work in hand. Precisely the same thing holds true of grinding wheels. It is quite as absurd to use the same stone for finishing brass and for sharpening tools; and likewise to use for grinding high-speed tools a wheel made for an entirely different class of work. If a stone be used which has been properly selected, and which is run under suitable con- ditions, there will be no glazing, even if the pressure be excessive. The latter condition will but tear up the wheel and overheat the tool the faster. Wheel for General Use. — It may be said, in general, that the difficulties just described arise from the use of a wheel too fine in grit or too hard (or close-grained) in bond. Wheels made especially for grinding hard carbon steel tools give but moderately good results when used on high- speed tools. The grain is too fine, and the grade slightly too hard. The ordinary run of high-speed tools such as are used in lathe, planer, shaper, boring mill, and the like work, require, for moderately rapid and sufficiently smooth grinding, a wheel of quite coarse grain. Mr. Taylor recommends for general work a mixture of grits, numbers 24 and 30; that is, grit passing through screens with openings respectively 24 and 30 to the inch. For all tools such as described above, unless intended for finishing cuts, a 20-combination is entirely satisfactory, and can well be used for all sorts of rough grinding, and even for a good deal of finish grinding. It is a mistake to suppose that to produce a fine finish in grinding the wheel necessarily must be of a fineness to correspond. This is true of soft metals, but not at all of very hard. The smoothness of the finish in this case, which includes high-speed steel, depends more upon the depth of cut (pressure applied), speed, and softness or open- ness of the wheel, than upon the fineness 1 of the grain. Wheel for Finishing Tools. — Tools requiring very keen cutting edges, like drills, milling cutters, and in fact all fine tools or such as are used for finishing cuts properly speaking, need a fine-grained wheel, as well as a softer one, to insure the best results — say what corresponds to a 60- grain J-grade alundum wheel. For tools intermediate in finishing 1 It may be mentioned here that the fineness of grain or grit is designated with approximate uniformity by makers of these wheels, by use of the numbers correspond- ing to the number of holes per inch in the screen used. For indicating "grade," how- ever, each maker seems to be a law unto himself and to use a different nomenclature — most frequently the letters of the alphabet. Even the very general terms "hard," "medium hard," "medium," "soft," and the like, vary more or less as applied by different makers. 138 HIGH-SPEED STEEL quality or size between the class just mentioned, and large roughing tools, a wheel of grit and hardness (or softness, rather, ) between this and that already designated for the roughing tools, is often used. A com- bination 20 and 30 is much in favor. It is a matter of great conse- quence that the wheel used be just suited to the tools; and for this reason it is very desirable that a sufficient variety, not only in form, but in grade and grain, be kept on hand, so that each batch of tools can be most eco- nomically and perfectly ground. The time required for changing wheels is in no wise comparable to that gained and the economy secured as a result of the changes. This raises the question of the organization of the tool-grinding service, which will be considered in another para- graph. To avoid the necessity of truing a wheel each time a change is made, it is desirable that each be mounted on its own arbor and screwed onto the spindle when required, or attached in some other way to secure perfect centering. ' Running Speed Important. — The speed of. the wheel should be that recommended by the maker. Not "somewhere near" it, but as closely approximating it as possible. This is very important, and the general disregard of it, the inattention to the maintenance of a suitable running speed, in general as rapid as the bond of the wheel will safely permit, is responsible for much of the trouble that arises in grinding. Indeed it may be said with assurance that practically all grinding troubles arise from ignorance of proper conditions, or inattention to them. Additional Considerations. — There are other conditions also involved in the proper grinding of high-speed tools — and of all classes of work likewise, for the matter of that. Evidently, for one thing, the wheel must be true; and for another, it must run steadily. To secure the latter condition, the machine, whether intended for hand or for auto- matic grinding, needs to be strongly built and even massive, and the spindles and bearings as large as may be consistent with the size of the wheels used. Provision of course must be made for keeping grit out of the bearings. It is a great mistake to suppose that any old worn-out stand will do, even for rough hand grinding. Truing Wheels. — The wheel once properly mounted (between soft metal- faced flanges with a diameter at least a third that of the wheel, or at any rate with suitable washers between wheel and flanges) in a rigid stand or machine, it must be exactly trued by a diamond dresser held firmly in position by the tool post or other holding device. Dressing by hand is unsatisfactory, and dressers other than the diamond do not yield a suffi- ciently good condition of the grinding surface. In truing, the diamond must be constantly and thoroughly flooded with water, or it is liable to be flattened. But little of the wheel material should be removed — only enough so the sound is absolutely even as the dresser passes back and forth over its face. Very light pressure, therefore, is required. GRINDING 139 Fig. 100. Dressing an emery wheel. This method is good enough for many kinds of work, but not for accurately grinding high-speed tools. Fig. 101. The dresser should be a diamond tool, held firmly in a post, as shown here. 140 HIGH-SPEED STEEL Automatic or Semi-Automatic Grinding. — Such precision as is here indicated of course implies automatic or semi-automatic machine grind- ing of tools, and is not essential in rough grinding by hand, though certain of the recommendations ob- viously apply in this case also. Hand grinding evidently has no place in a well-regulated shop manufacturing or using tools in such quantity as to warrant an adequate equipment for putting and keeping them in proper shape. Nothing is more certain than that a large part of the ineffective Fig. 102. Walker self-contained floor grinder. This grinder has much to commend it where hand grinding is permissible. The rotating hood forms also a bowl which may be kept filled with water, for the collection of dust. Caliper ■ests, as here shown, are essential in the hand grinding of high-speed tools. Fig. 103. Yankee drill grinder. Motor-driven and entirely self-contained. It is essential in a drill grinder that the holder be so swung as to grind with correct clearance, as this one does. work of tools, and the frequently large loss by breakage common in so many shops, is due to improper grinding. A drill with lips ground at a guess, one lip sure to be different from the other, as is unavoidably the case when it is ground by hand, even by an experienced workman, GRINDING 141 clearly goes into its work with conditions favorable to breakage and with every probability that the holes produced will be imperfect. Mill- ing cutters, lathe tools, and the like, of course are from their forms less likely to breakage under strains; but inequalities in cutting edges, espe- cially so in the teeth of milling cutters, core reamers, and the like, inequalities inevitable in hand grinding, very evidently will show up in the surfaces they leave behind. Furthermore, they tend to promote chattering. It should be obvious, therefore, that hand grinding has no place in any well-regulated large shop, except possibly for roughing tools down to approximate size, and that the precision above recom- mended is none too great to insure the highest efficiency in high-speed tools. Grinding Equipment. — As to the number and kind of machines to be installed, this naturally would depend largely upon the quantity and Fig. 104. How the clearance of a drill increases from the periphery toward the center. Suppose A and B represent cylinders corresponding in diameters to any two points in the cutting lip of a drill, and c d the feed or advance per revolution, for the sake of clearness much exaggerated. The angles made by the lines 6 and a then represent the clearance required at the selected points. Courtesy of Wil- marth & Mormon Company. kinds of tools used. In a shop of any considerable size it is likely that at least one drill grinder would be required, preferably one easily adapted to the grinding of all sizes used, unless there be work enough to keep more than one machine busy. Most drill grinders now offered conform to the two prime essentials — freedom from vibration, and adjustment for maintaining uniform lip angles and curves at the points, for all sizes to be ground. It is of the utmost importance in grinding drills that the clearance angles along the entire length of the cutting edges be uniform; otherwise the clearance at some points will be too great, and at others too slight, as is shown by the annexed Fig. 104. In order to accomplish this, the drill holder must be so devised as to swing through a curve corresponding to the required clearance angle. 142 HIGH-SPEED STEEL Since lathe and like-shaped took form by far the larger portion of the tools in most shops, a universal machine adapted to grinding these tools is generally essential. Milling cutters can be ground successfully only in a machine designed for the purpose, or by the use of attachments to other grinders providing the requisite fittings and movements. In small shops such a combination machine, say like the one illustrated at Fig. 107, is well calculated to take care of all kinds of work. Some of the more expensive universal grinders likewise are now provided with attachments permitting the grinding of other than rotary tools (Fig. 107). Fig. 105. A universal grinder (Gisholt) designed for sharpening lathe and like tools. All angles can be produced with certainty, and by following a chart giving the standards adopted, tools can be redup- licated indefinitely. It is sometimes desirable to make use of special grinders for special forms of tools, whether because of the amount of work to be done or the superior adaptation of such machines to the work thus in hand. An example of such special machine is a saw grinder and a special reamer grinder, illustrated at Figs. 108, 109 and 110. Special devices in the nature of jigs can be used to a much greater extent than is now done, in grinding not only inserted cutter blades, but other tools and parts. Cup wheels, used in common with disc grinders on other forms of tools, are required for most rotary cutters, since this shape of wheel allows the proper facing and backing off of spiral and other difficult shapes of cutters. The cup wheel gives a straight clearance or land instead of a GRINDING 143 Fig. 106. Wilmarth & Mormon Company yankee drill grinder, with attachment for universal grinding. Very desirable in a small shop with a considerable range of tool grinding work. Fig. 107. In shops using relatively few tools, a grinder resembling this Brown & Sharpe machine is very desirable, since it is available for practically all classes of tools, including those of the lathe and planer 144 HIGH-SPEED STEEL Fig. 108. A special grinder for the Tindel inserted tooth metal saw. Saves frequent readjustment of a universal machine used for other purposes. Nft^Sfi ■ ^ m l%Jr m&» V* V * ^» ■ 1 I 1 1 ' Ki ^ 13 . .^^^r Fig. 109. Sharpening a Tindel saw on a Le Blond universal grinder. GRINDING 145 Fig. 110. Grinding a rose reamer on a special machine. This is the economical method of doing such work where many tools of the kind are used. Fig. 111. Grinding jig designed by William G. Thumm. Especially adapted for grinding inserted cutters for a large face mill. 146 HIGH-SPEED STEEL curved one, such as is obtained (unless unusual precautions are taken) when tools are ground on the periphery of a disc wheel. This method of grinding, by the use of cup wheels, therefore not only does not undercut the edge, but leaves it in the best possible form and condition for effective work and maximum life per grinding. Fig. 112. The cup wheel has decided advantages in certain kinds of work, as in edging a shear blade, requiring a flat land or surface behind the cutting edge. Conditions to be Avoided. — Some grinding machines are provided with positive feed devices for forcing the tool against the wheel. There is no objection to this arrangement if the feed be light, as already recom- mended, and if provision be made at the same time for moving the tool GRINDING 147 £*,_. T \f / or wheel in such a way that one passes across the face of the other to a greater or less extent during the whole time of the grinding. If tool and wheel face maintain the same rela- tive position, even with a light feed, the chances are that they will quickly come to fit against each other very closely. The cutting face of the wheel then gets smooth, the grinding proceeds slowly or entirely ceases, and the tool rapidly heats up, just as if the wheel were glazed — which it sometimes is under these cir- cumstances. Such a condition is most likely to occur when the face of the tool is rather large, and in this case especial care is to be observed when / grinding With a flat Surface. Grinding Fig. 113. Effect of grinding with disc and ., » .., ., , . with cup wheels. across the lace with the angle of a wheel having a V-shaped instead of a band periphery, eliminates this trouble, though perhaps it somewhat reduces the rapidity of the work and at the same time leaves a face more or less curved according to the diameter of the wheel, as is the case always in grinding on the Fig. 114. Sellers grinder for flat-face tools. The wheel with a V-shaped periphery has certain advantages. 148 HIGH-SPEED STEEL periphery of a wheel. The method makes it possible to flush thoroughly, a difficult thing to do in flat grinding. When a relatively large surface presses against a wheel surface, very little, if any, fluid gets between; so that the purpose of flushing is in large measure defeated. Wet vs. Dry Grinding. — As to the respective merits of wet and dry grinding it does not seem safe to hazard a square statement. Many have found, or think they have found, wet grinding advantageous; and many others seem to have a contrary experience. The purpose of wet grinding of course is to cool the tool and therefore to allow more rapid work; and incidentally to eliminate dust. With suitably hooded ma- chines the dust is effectually removed; and it is a question if the damage often done tools in wet grinding does not much more than offset the possible increased speed. If the amount of water thrown against the tool is very large and the stream is closely confined to the place where the work is done, there is considerable gain in large work. There is, however, the practical difficulty of forcing water between wheel and tool surfaces where it can keep the face cool, all the greater because the amount should be large, but the speed of delivery slow; and it is doubtful under these circumstances if the liability to checking by reason of the contact of the water with the at times over-heated sur- face does not do more damage than good. However that may be, wet grinding is very largely practiced in connection with large and simple tools, especially where the surfaces to be ground are more or less round- ing rather than flat. On such work as milling cutters, reamers, drills, and the like, dry grinding seems preferable; and indeed few machines are designed for wet grinding of these types. The sandstone, if used, must run wet; and this is a good reason why it is best left alone for much, if not all, grinding of high-speed tools. Oil for Cooling — Nozzles and Hoods. — Modern emery or composition stones are not affected by oil, when well soaked, and do not, so far as reported, have their cutting qualities impaired. It would therefore seem that if oil were used for flushing in grinding all classes of high speed steel tools, all the advantages of wet grinding would be obtained, with none of its disadvantages. The very considerable waste from " spattering " could no doubt be eliminated by suitably constructed nozzles and hoods. All grinders should be hooded anyway; and as for nozzles, until recently nobody seems to have thought it worth while to use anything other than a piece of pipe. With some attention to these points oil grinding would seem to promise much in this new' field. Grinding Before Hardening. — Whatever may be the several opinions respecting sandstone and emery wheels, and dry or wet grinding of hardened tools, the kind so far under consideration, it seems to be pretty generally agreed that for pre-grinding, that is, for grinding before harden- ing, a dry emery wheel is most satisfactory. With a soft or open and GRINDING 149 coarse wheel, material can be removed with great rapidity . and without danger of harm to the tool. Sometimes the tool is ground while still hot from the forging heat; and indeed this is recommended where con- venient. It is no disadvantage even though the tool be still red hot. The advantage ofthuspre-grind- ing manifestly is that the tool is still soft, and the reduction is much more rapid than after it has been hardened. Metal to be Removed in Grinding. — The finish grinding obviously need be relatively slight. This re- fers to the sur- face of the tool in general; for at the cutting edge the finish grind- ing, when rough grinding pre- cedes the hard- ening process, must be severe, if the full effi- ciency of heavy tools is to be de- veloped. It is a matter of remark among the users of high-speed tools that in general they do not work up to their highest possibilities until after two or more grindings. The reason is simple. The high hardening temperature to which they are subjected affects the sustaining power, especially at the cutting edge, where the danger of " burning " the steel is greatest. Evidently a tool will dull or break down much more rapidly when any portion of the injured skin remains than when this has been removed; and quite evidently also the tool must either be ground several times in the customary manner, or the burnt portion must be removed by a severe single grinding, in order to bring out its full possi- Fig. 115. The " Wizard nozzle prevents the spattering common when ordinary- forms are used. 150 HIGH-SPEED STEEL bilities at the first use. On lathe, planer, and similar tools, T V mcft is ordinarily none too much to grind off the cutting edge the first time; and on large tools rather roughly forged, more is desirable. On fine cutters with keen edges, hardened at a somewhat lower temperature, no such heavy first grinding is necessary. Ordinarily such tools are set at work without grinding subsequent to hardening. Direction Wheel Should Run. — Carbon steel tools are ground with the wheel running toward or from the cutting edge, at the fancy of the grinder usually ; and good edges can be secured either way — provided, in the second case, the burr be removed by honing. High-speed tools are preferably ground with the wheel running against the cut- ting edge, most grinders being in this way able to get better results. It is not desirable that the grinding face be run along a cutting edge, though in certain cases this may be unavoidable. Revolving cutters, when so ground, that is, with the wheel rotating against the cutting edge, need Fig. 116. High-speed tools are best ground by re- volving the wheel against the cutting edges, as show in B above, rather than with the teeth, as in A. The former method prevents burring and allows faster grinding. The cutter must be held firmly against the rest, by hand or otherwise. Fig. 117. Customarily the cutter is held by hand against a guide. Especially when used in taking finishing cuts, it is better that the cutter be held in place mechanically. to be rigidly held against the tooth rest; otherwise there is likelihood that they may be drawn out of proper position by the wheel, and the teeth scored and the wheel damaged or even broken. The usual method GRINDING 151 is to hold the tool by hand against a guide. This is liable to permit more or less unsteadiness and consequently eccentricity in the periphery of the cutter. Holding and rotating the cutter mechanically is much to be preferred, particularly if it is to be used for finishing cuts. Amount of Water Required. — It seems scarcely necessary to point out that a tool started either wet or dry should be finished without change. In hand grinding dipping the partly ground tool into water for cooling is almost sure to damage it. In wet grinding the water supply must be much more liberal than is usually allowed. The flow need not, and indeed should not, be very rapid; but the volume delivered from the nozzle must be, for ordinary grinding, enough to flood completely the tool — say from five to ten gallons per minute. A discharge area equivalent to that of a f inch pipe, therefore, is none too large, and for big tools is not large enough. Keeping Tools Sharp. — Because it is possible to force high-speed tools even when dull, it is a common practice to run them longer without re- grinding than is economical — to run them, in fact, until the edge breaks Fig. 118. Grinders in connection with the stock room. This is a very convenient location for grinders, the work being done by those connected with the tool or stock room. down or the product passes beyond the limits of required accuracy. The maxim " keep your tools sharp " is as applicable in the case of 152 HIGH-SPEED STEEL high-speed as in that of ordinary tools; though the consequences of disregarding it are perhaps less noticeable when the tools are used in connection with strictly modern machine tools. If used in machines not especially designed for them, the observance of the caution is a matter of great importance; otherwise the wear on the machine and effect upon the work is very marked. Furthermore, if the tool is not ground as soon as it begins noticeably to dull, the dulling thereafter proceeds at an increasingly rapid rate on account of the machine giving under the increased strains and the consequent augmentation of the chatter, which latter is at the bottom of most of the wear or breaking down at the cutting edge. In the end, therefore, it is better to grind oftener, remove less metal per grinding, and keep the tools keen. This will almost wholly obviate the need for fettling tools in the forge shop, particularly if they have been properly designed in the first place to provide for the removal of many successive layers at the cutting end before requiring forging to shape again. Re-grinding — Finding Cracks. — In re-grinding tools, either because dulled or because of damage sustained in a previous grinding, the amount of metal removed should be commensurate with the condition of the cutting edge or tool surface. If the tool has been damaged by over- heating in grinding, the part to be ground away most generally will need to be at least T V inch, and may need to be two or three times as deep. Checks are exceedingly difficult to discover, and usually pass unnoticed until the tool breaks down at work. Mention is made hereafter of a method whereby they can usually be detected. Land vs. Face Grinding. — When re-dressing a dulled tool it is desirable, for the most part, to grind both lip or face, and clearance or back, most of the material removed preferably coming from the latter surface. As tools of the milling cutter type usually wear, a given depth removed from the back will give a result equivalent to that produced by remov- ing two or three times as great a depth from the face. Grinding the back also serves to preserve perfectly the contour of the cutting pe- riphery. The life of such cutters is therefore considerably prolonged in this way. Tool-Room Organization. — The methods and precision here indicated as essential to economical grinding for highest efficiency, imply the standardization of the tools used in a shop, as far as possible; and the organization of the tool supply on a basis which relegates all grinding to a department or to departments suitably equipped for first-class work, and manned by operators skilled in that especial work and trained to the observance of all details of design in particular tools as well as the methods to be followed in the actual grinding operations. The organiza- tion of a tool-supply department can be very simple, and indeed should be so. The prime requirements are proper equipment and operation, GRINDING 153 Fig. 119. The difficulties formerly attending the accurate grinding of spiral milling cutters have been pretty well eliminated in several recent universal cutter grinders. A cup wheel grinding the land back of the cutting edge. Fig. 120. An expensive method of grinding high-speed or any other tools, viewed from whatever point. 154 HIGH-SPEED STEEL ROUND NOSE ROUGHING TOOLS i SillERS AG° Imcowi* for LATHES * PLANERS. u:-. -. -,r NO "••■'- SuPinstDiht No 19598 Blunt Tools, roR Cast Iron * Haroer Grades of Steel Sharp Tools, for Wrought Iron a Softer Grades of Steel. To Grino TOP FACE AojuSt Machine as Follows: To Grind TOP FACE Aojust Machine as Foll ton STRAIGHT TOOLS. for STRAIGHT TOOLS. RIGHT HAND LEFT HAND. RIGHT HAND. LEFT HANO. Horizontal Ancle 97.°7 Horizontal Anclc 97.7: Horizontal Ancle 975 Horizontal Anclc 37% 104' Veatical Ancle Vertical Anc^e 112' Vertical Ancle for BENT TOOLS. for BENT TOOLS. RIGHT HAND LEFT HAND. RICHT HAND. LEFT HANO. Horizontal Ancle I03!7 Horizontal Ancle Horizontal Ancle 107'; Horizontal Ancle 107.'; Vertical Ancle Vertical Ancle Vertical Anclc I05T6 Vertical Angle To Grind END FACE Use FORMER (A) and Make Horizontal Ancle or Adjustment 86 When Face is Finished Index Finger of END GAUGE Should Point as Follows: rpr Size or Tool \ y z Size of Tool >¥' Inoc» Finccr Should Point to II 15 27 Index Fincfr Shoulo Point to 10 Sthaicmt Tool Ric>u Hand Straight Tool Lctt Hano. ffiONT R»*t^ _V Bent Tool Richt Hand Bent Tool Left Hand. Fig. 121. Part of direction sheet used in connection with Sellers' grinder. With a sheet thus showing standard forms and angles before him, the tool grinder is able to reproduce exactly a tool edge an indefinite number of times. as stated; standard designs of tools for practically all jobs, all details for each tool being definitely determined and carefully observed; and an ample supply of tools, sufficient to permit machine operators to replace dull tools without loss of time. The latter are at convenient times returned to the supply department in exchange for sharp ones. The tools are ground in batches, as these accLimulate, to save too frequent changes at the grinder. The man in charge of the grinding, of course, is provided with a set of standard samples and a chart indicating the precise form and angles for each tool. This being carefully followed, the difficulties arising from hand grinding, such as varying angles, unsymmetrical cutting edges, improper backing off and relief, and the like, are entirely absent, and tools not only work more effectively but last a great deal longer. The fixed charges on the investment repre- sented in such an equipment and ample tool supply is not comparable to the economy effected. The conditions of course apply with equal force in the case of ordinary tools; and the method of handling the re- grinding would be the same for both. CHAPTER XII. DETERMINING AND REGULATING TEMPERATURES IN HIGH SPEED STEEL TREATMENT. Reproducing Determined Conditions. — Guesswork is not consistent with modern industrial methods. Rough approximation is uncertain, and therefore wasteful. It has, of course, always been true, but only in recent years has it come to be well understood, that the physical and chemical changes involved in so many productive industries take place under definite and constant conditions. Variation in conditions, whether it be in burning coal under a boiler, conducting an electric current, the heating of a baking kiln, the treatment of a tool, or what not, involves variation in the nature or efficiency of the product; and in consequence also it involves waste. The conditions of maximum effect once definitely determined, it is of the first importance in nearly all industrial operations that they be reduplicated within the established limits, with certainty and economy. The manufacture of high speed steel tools forms no exception; on the other hand the accurate gaging and reduplication of temperatures, especially high temperatures, is an absolute essential to anything approaching the maximum efficiency in tools. The Eye not Dependable. — Formerly a prime qualification of a success- ful tools mith was the possession of a well-trained eye, the ability to dis- criminate sharply the colors through the wide range seen in the heating of a piece of steel — this, that he might gage with more or less accuracy the heat to which the tools of various steels and for different uses were to be raised when hardening or tempering. Not that the color scale had for him (usually, at any rate,) any definite relation to specific tempera- tures, but rather because it was known to be more or less definitely related to the hardness and lasting quality of steel tools, the relation depending a good deal upon the particular steel used, and perhaps also upon other conditions. No matter how skillful a toolsmith might be, however, his tools, carefully made as nearly uniform as might be, still turned out varying more or less in quality. As we now know, this of course is just what might be expected under the circumstances. Until lately nothing was known of the critical or recalescence points in steel, the precise location of which in the temperature scale must be known before the proper heat for any particular steel can be determined. Even 155 156 HIGH-SPEED STEEL if they had been known, their precise location by reference to the colors as perceived in a piece of steel by the unaided eye, would have been practically impossible. This is due not only to the difficulty of discrim- inating between the colors of a radiant body when those colors do not vary greatly, but even more to the personal equation of the observer and the variation in the conditions under which the observations are made. Elements Making for Uncertainty. — Not only do persons differ as to just what is, say bright red or light yellow, or any other color for the matter of that; but in the same person the judgment will vary with his freedom from fatigue, his physical condition, and even his mood. Furthermore, the light in which colors are seen modifies them to a con- siderable extent, so that seen in one part of a shop a piece of steel of a given temperature might appear to have one color, while in another part of the same shop it might appear several shades different. Even in the same spot in a shop the light is bound to vary a good deal accord- ing to the cleanness of the window and the condition of the weather outside. Of course when artificial light is necessarily used part of the time, the trouble is still further accentuated. The Personal Equation. — The difficulties arising from the personal equation, even in the case of skilled observers, is well seen in a com- parison of three well-known color scales: TABLE VI. _^ u _^ u • t-i Pouillet. 1 a CD S3 Taylor & White. a £3 Howe. c Q3 si O fa O 566 fa 1050 u fa Lowest red Dark red 525 700 977 1292 Low, dark, or) blood red J Lowest red vis-) ible in dark j Lowest red vis- ) ible in light ) Dull red | 470 475 550 to 878 887 1022 to Lowest cherry 800 1472 Dark cherry 635 1175 I 1 625 1157 Cherry 900 1652 Cherry, full red 746 1375 Full cherry 700 1292 Bright cherry 1000 1832 Light red 843 1550 Light red 850 1562 Dark orange 1100 2012 Grange 899 1650 Bright orange 1200 2192 Light orange 941 1725 f 950 1742 Yellow 996 1825 Full yellow < to 1000 to 1832 Light yellow 1079 1975 Light yellow 1050 1922 White 1300 2372 White 1205 2200 White 1150 2108 Bright white 2 1400 f 1500 2552 2732 Dazzling white \ to [1600 to 2912 1 Pouillet over seventy-five years ago devised his color scale, which even to this day is quoted as authoritative, though his instruments for gaging temperatures were by no means comparable with those in use to-day, and his results, or his nomenclature, at any rate, are not very well in accord with more recent scales, as may be seen in the table above. 2 It is impossible, in tool-making practice, to discriminate with any accuracy the hues generally designated white. REGULATING TEMPERATURES IN HIGH-SPEED STEEL 157 Dependence upon the eye for the determination of temperatures, and their accurate reproduction, leads to very uncertain, and in the case of high-speed steels, rather unsatisfactory results. It is quite as evident that in the treatment of a commercial product as expensive as a compli- cated tool, particularly when made of a material like high-speed steel, which requires for maximum effects even more care than the cheaper carbon steel tools, adequate means for temperature gaging and redupli- cation are of prime importance. Temperature Gaging Devices. — A good many instruments and devices have been brought forward for gaging temperatures, especially tempera- tures above those for whose measurement the spirit of mercuric ther- mometer is available. Those at present in use, which are of value in connection with industrial processes, are included in the conspectus here presented (Page 158). Adaptation of Gage to Purpose. — Not all the instruments included in the table (VII) are suitable for use in connection with the treatment of steels, and some of them can be used to a limited extent only. Thus the mercurial thermometer, when of suitable form, can be used for gaging the temperature of the oil tempering bath. When so used it needs to be made of specially heavy glass, well annealed, and preferably the tube above the mercury filled with an inert gas under great pressure. " Sentinel " Pyrometers, or Temperature Cones. — " Sentinel " pyrom- eters, or temperature cones, strictly speaking, are not instrumental, for Fig. 122. Temperature determining cones, or " sentinel " pyrometers, which melt down or fuse when predetermined temperatures are reached. there is no scale, and each cone can be used but once. They are made of metallic alloys or mixtures or earths and the like, so proportioned that when a given cone reaches a predetermined temperature it fuses or melts down. They are useful, therefore, in indicating when desired 158 HIGH-SPEED STEEL TABLE VII. Class. Principle Involved. Types. Range Centigrade. Range Fahrenheit. Expansion Variation in volume Gas 0—1000 32—1850 of a substance by Mercurial -C4— 550 -100—1025 change of temper- Spirit -210— 25 -350— 75 ature. Pneumatic, gas low— 980 low— 1800 Metal rod, etc. low— 485 low— 900 Porcelain (contrac- tion) low— 1800 low— 3200 Water current 0—1600 0—2900 Brown platinum 0—1600 0—2900 Pneumatic Flow of gases thru small apertures. Uehling low— 1650 low— 3000 Calorimet- Relation of specific Siemens water py- ric heat to quantity absorbed. rometer low— 1480 low— 2700 Fusion or Unequal fusibility Seger temperature "Sentinel" of various metal- lic or earthy cones cones 0—2000 32—3600 Thermo- Current developed LeChatelier \ electric when one junction of a thermo-couple is at a temperature differing from that of the other. Hoskins [ Bristol C Price, etc. / -180—1650 -380—3000 Resistance Variation in electri- LeChatelier lowest 1 lowest 1 cal conductivity attain-|l200 attain-|2200 under changes of able J able J temperature. Radiation Measurement of heat radiated. Fery ) Bolometer \ 900— 1600 2 1650— 2900 2 Optical Variation in luminos- ity or wave length. Morse — superposi- tion of incandes- cent filament. 600—2000' 1100 — 3600 1 Le Chatelier mirror Fery absorption, I 500— 2000 1 925— 3600 1 Wanner — all pho- 900— 1800 1 1650— 3250 1 tometric compar- ison. Mesure & Nouel — prismatic 750— 1000 1 1400— 1850 1 1 The upper limits of some of these pyrometers are much higher than those here indicated — in some cases, as the Fery radiation instrument, and the bolometer, there is no theoretical higher limit. The ranges here given are, so far as data have been obtainable, those within which reasonably ac- curate results are to be had in industrial service. Certain of the instruments named are sometimes made for laboratory use with higher and lower limits. The Holborn-Kurlbaum instrument is the German form of the Morse, and its range is the same. 2 The bolometer, especially in its improved form, is a laboratory instrument capable of measuring infinitesimal changes in temperature, and has an unlimited theoretical range. A change as minute as the millionth part of a degree can be measured with it. In the improved form it consists essen- tially of a pair of differential platinum thermometers made of very narrow strips of exceedingly thin foil, one of which is completely blacked and the other bare, both enclosed in an hermetically sealed tube. An indicator of the Callendar type is generally used. REGULATING TEMPERATURES IN HIGH-SPEED STEEL 159 temperatures are reached, and can be made to indicate, by using pairs selected for the purpose, the maximum and minimum temperatures within which a process or treatment must be carried on. Thus if a tool is to be heated to between, say 1025 and 1050 degrees C, two cones, one of which fuses at the lower temperature and the other at the higher, are placed in the furnace while the heat is going up. When the first cone melts down, the tool is introduced. By watching the second cone and regulating the furnace when there are signs of its melting, the tem- perature can be maintained within the required limits. It is well to introduce cones of the first kind from time to time so as to insure keeping the temperature above the minimum. This method of course is rather tedious, and is not permissible as a regular thing in commercial work. The " sentinels " are very useful as a check upon the regular pyrometer, and also upon the judgment of the operator when tools are hardened without such an instrument. Some sixty different grades are com- mercially obtainable, giving a considerable range, from about 590 to 1975 degrees C. (1095 to 3720 F.). " Poker " or Fire-End Pyrometers. — The most usual method of gaging temperatures of materials undergoing industrial processes is to insert into the furnace, fire chamber, or other containing receptacle, the so- called fire end or " poker " of any one of several types of pyrometers, the temperature being indicated or recorded at a greater or less distance away (in the case of the electrical instruments), as may be expedient. In expansion instruments the indicators of course are attached directly to the stem, as in the water current and the Brown platinum pyrometers. These instruments are based on the same principle, the former having its non-platinum parts cooled by a current of water so as to adapt it for continuous use. The latter is quite durable, but is not adapted for continuous use, the stem or fire end being exposed to the furnace heat only long enough to allow the indicator to register the maximum, the instrument being withdrawn before the iron frame can be injured or the platinum impaired. Le Chatelier Type. — Le Chatelier seems to have been the first to per- fect an electrical heat gage, applicable to industrial processes requiring very high temperatures, in his thermo-couple electric pyrometer; and instruments of this type are most frequently used in connection with the treatment of high-speed steels. Several different makes are on the market, varying in excellence, reliability and endurance. All, however, are based on the principle that when one junction of a thermo- couple (that is, of a continuous circuit composed of two different kinds of conductor) is heated while the other remains at a constant normal temperature, a feeble electric current is set up which varies more or less regularly according to the materials constituting the thermo-couple, with the temperature to which the " hot " junction is exposed. The 160 HIGH-SPEED STEEL current of course can be measured by means of a delicate galvanometer, and the relations between the strength of current and the temperature Fig. 123. Le Chatelier fire end inserted in oven furnace. The indicator is at any convenient distance. Fig. 124. Brown quick-acting platinum pyrometer, for intermittent service. of the junction having been determined, the deflections of the galva- nometer needle may be converted into temperature indications. Owing to the " extra currents " and other disturbing elements found in pairs of most materials otherwise suitable for the purpose, and likewise REGULATING TEMPERATURES IN HIGH-SPEED STEEL 161 because of the necessity for elements capable of withstanding extremely- high temperatures, the thermo-couple material must be selected and manufactured with extreme care. Most often, for the determination of rNTERCHANOEABtjE MOUIAfl OH tTIUIOHT CONNECTION /S)T TERMINAL BOX COVER WIRE-CONNECTOR Fig. 125. StupakoS (Le Chatelier) pyrometer outfit. Courtesy of Charles Engelhardt, New York. such high temperatures as are involved in high speed steel treatment, the couple is constituted of platinum and a platinum-rhodium alloy, though nickel-chrome steel and other nickel alloys also are used. Maximum Temperature Range. — The heat resistance of these elements is very high, making it possible to expose the instrument, when made Fig. 126. Thermo-couple pyrometer enclosed in steel tube, and indicating instrument. in the best form, to temperatures up to 1600 degrees C. (2920 F.) and even above, for short periods. This is considerably above any tempera- ture required in the making of tools, 1400 degrees C. (2550 F.), or there- abouts, being the highest usually required. At these temperatures, 162 HIGH-SPEED STEEL however, the fire ends deteriorate with greater or less rapidity, and are easily broken afterward. They are, in order to prolong their period of accuracy and permanency, usually protected by being enclosed within porcelain, fire clay or other heat-resisting material (a combination of asbestos and carborundum, in the case of the Bristol instrument); but these frequently crack and crumble, and it is necessary to check the instruments at intervals against standards of known certainty, to in- x^ -^ INEXPENSIVE SUBSTITUTE FOR PLATINUM-RHODIUM COUPLE Fig. 127. Bristol compound fire end. sure consistency in the readings, if not accuracy. It is shown hereafter that absolute accuracy, that is, the indicating of the absolute tempera- ture, is less essential than consistency; and so, if an instrument has departed from its calibration, the variation being within reasonable limit, allowance can be made for this, provided the extent is known. When the departure has reached an amount where the indications no longer are reliable, of course the instrument must be discarded. Fig. 128. Hoskins pyrometer with exposed couple. Recent Thermo-Couple Developments. — A recent development in py- rometers of this type consists in making the fire end compound (Fig. 127), REGULATING TEMPERATURES IN HIGH-SPEED STEEL 163 only that portion actually exposed to the high temperature being of the precious metals. This reduces the expense of renewals considerably, while the readings are sufficiently accurate for the present purpose. Fig. 129. Hoskins standard fire end, standard thermo-couple with handle and leads, and new nickel sheath thermo-couple complete! The fire end is made of heavy alloy wires twisted and welded. Another development is the Hoskins thermo-couple, which is made of comparatively inexpensive alloys that withstand the required high heats Fig. 130. Application of poker to oven furnace, and Bristol method of cold end temperature maintenance. apparently indefinitely, while at the same time they are made of wires heavy enough to require no protection and to allow of considerable rough usage without detriment. The fire end junction being exposed 164 HIGH-SPEED STEEL directly to the temperature to be measured without the intervention of protecting tubes or covers, responds quickly, and there is little or no lag in the indication. A modified form of this instru- ment has one element in the form of an asbestos- wound wire enclosed within and joined to a nickel tube which forms the other element. Protection of Fire Ends. — It is well enough, and in the case of most of these instruments necessary, to provide an iron pipe covering, perhaps of ordinary gas pipe, closed at the inner end so as to form a well, not only for the protection of the fire end, but to prevent unnecessary and often detrimental air currents. Cold End Temperature Compensation. — Inasmuch also as the correct registering of the temperature under observation depends upon the "cold" end of the couple, that exposed to the normal atmospheric tem- perature and near the temperature for which the in- strument is calibrated (usually about 25 degrees C. or 75 F.), that end should be out of the range of direct radiation of the furnace, or other sources of tempera- ture variations. The Bristol instrument has an ar- rangement whereby the cold end is kept near the Fig. 131. Section of an , . , electrical resistance floor; and in case still greater accuracy is required than is thus afforded, a compensator is placed in the circuit to preserve the calibration practically correct. Unless the atmos- pheric temperature varies considerably from that above given (25 degrees C), allowance can be made for it in reading the indicator. This allowance is governed by the design of the instrument, and is usually furnished by the maker, with the instrument. At least one indicator on the market utilizes a multiple scale whereby the allowance is made automatically by reading the scale nearest corresponding to the atmospheric tempera- ture, as shown by a mercurial thermometer provided for the purpose. Theoretically the temperature of the cold end should be at zero; and in laboratory work, and in certain industrial operations where much refinement is necessary, an " ice bobbin " is utilized for maintaining this condition. In ordinary industrial processes, such as steel hardening, where the length of the elements is sufficient to allow the cold junction to be moderately near the limits already indicated, the slight variations due to this cause may be disregarded. Electrical Resistance Pyrometer. — The resistance type of electrical pyrometer is also used to some extent in hardening high-speed steels, and for all temperatures below 850 degrees C. (1600 F.) is perhaps the most accurate type, suitable especially where a very open scale is desired. It depends for its action upon the variation in the electrical conductivity REGULATING TEMPERATURES IN HIGH-SPEED STEEL 165 of a platinum wire or foil, according to the temperature to which it is exposed. This variation is practically constant, and when measured by an indicator constructed on the principle of the Wheatstone bridge, can be easily reduced to temperature units — or most usually, read off directly on a temperature scale to which the indicator has been calibrated. The indicators and recorders are cumbersome and much more expensive than those for thermo-couple pyrometers of the same range. The latter are better adapted to measuring high speed steel hardening temperatures because the resistance instru- ments will not stand exposure to the intense heats required, except for very short periods, the maximum for such short periods even being only 1200 degrees C. (2200 F.). For that matter, it is desirable that neither kind be exposed unnecessarily; and especially when enclosed in porcelain or similar protective cylinders they must be heated up rather gradually. The fire ends crumble and deteriorate rapidly enough with careful usage. Besides being unsuited for continued use at temperatures above 850 degrees C. (1600 F.), the cost of the resistance fire ends, as well as of the indicators or recorders, is very much higher than that of the thermo-couple type. Water Current and Uehling Instruments. — Two other pyrometers have been used in connection with hardening high-speed tools, with good results — the water current and the Uehling pneumatic. The latter of these, however, is very expensive and has no particular advantage over the electric instruments already Fro. 132. Le Chatelier resistance pyrometer. Fig. 133. Mesure and Nouel optical pyrometer. O, ground diffusing glass; P, polarizing nicol; Q, quartz plate; A, analyzer; OL, eyepiece; K, rack and pinion; C, graduated circle, calibrated and reduced to temperature scale; L, objective. described. Both have all non-platinum parts which are exposed to the fire, cooled by the circulation of water through or around them. Their accuracy and permanency, therefore, is much greater than that of most other forms of fire and temperature gages. Fire-End Deterioration — Optical Pyrometry. — The deterioration of the fire ends is the weak point in most pyrometers of that type; and to obviate the difficulty instruments have been devised which do not 166 HIGH-SPEED STEEL require having any part directly exposed to the high temperature sought to be gaged. These, with one exception, are of the optical type; and all utilize the energy of radiant matter transmitted to any convenient distance, in the determination of the temperature of the body under observation. Mesure and Nouel Pyrometer. — Of all the pyrometers adapted to use in hardening operations, the Mesure and Nouel (Fig. 133) is perhaps the simplest, since it is entirely self-contained and has no delicate parts to get K Fig. 134. Fe"ry mirror or absorption pyrometer. DB, a small telescope with B the eyepiece; E, standard lamp; F, mirror; CCi, absorbing wedges. out of order. It utilizes the colored field produced by the polarization of light from the object observed, and the accuracy of a temperature reading or of its maintenance depends upon the judgment of relative colors, very much as when the natural colors of a heated object are viewed by the unaided eye. For this reason, among others, it is of material assistance only in the hands of a skilled operator, and even such an one cannot be sure of any accuracy within fifty or more degrees C. at temperatures above 1000 degrees C. Photometric Type of Instrument. — The LeChatelier and the Wanner optical, and the Fery absorption pyrometer each depends upon a photo- metric comparison of the relative brightness of the two halves of the Fig. 135. Wanner optical pyrometer. A, Nicol analyzer; B, biprism for eliminating images; D, slit through which images are observed; E, eye- piece, 2 , lens for focusing image upon.D; O,, objective; P, d-v prism; R, Rochon prism; Si, slit for admis- sion of light from standard; S^, slit for admission of light from object observed. illuminated field, one half receiving its light from the radiant object, the other half from a standard source of light forming- part of the instrument. The Le Chatelier instrument utilizes an iris diaphragm for regulating the amount of light admitted from the radiant object, in combination with a mirror and a standard source of light of known intensity, the light from the two sources each covering half the field. By adjusting the diaphragm until the two halves are of equal bright- REGULATING TEMPERATURES IN HIGH-SPEED STEEL 167 ness, the temperature can be calculated, or read off directly, from the scale attached to the diaphragm. Fig. 136. Le Chatelier optical pyrometer. Fery uses a system of absorbent wedges for the same purpose, and the reading is taken in the same way. The Wanner instrument utilizes, in combination with the standard source of light, a system of prisms and lenses for polarizing in such a way that by turning the analyzer with its attached graduated scale, the two halves of the illuminated field, one receiving light from the standard of comparison, and the other from the object observed, may be made of the same luminosity and the temperature then read off at the scale. All the above pyrometers, except the Mesure and Nouel, are quite accurate in the determination of relative temperatures within their several ranges. Generally speaking, the possible error is less than one per cent, and in some cases only half as great. Morse Thermo Gage. — Perhaps the most convenient of the optical pyrometers which are accurate enough for use in connection with high speed steel treatment is the Morse Thermo Gage, made in Europe with some slight modification under the name Holborn-Kurlbaum Pyrometer. It consists of a tube, furnished with lenses if desired, within which is the filament of a small low- voltage electric lamp. In the lamp circuit is a rheostat and a milliammeter. In determining the temperature of a radiant substance it is observed through the tube, and the current passing through the filament is at the same time so regulated through 168 HIGH-SPEED STEEL the rheostat that the filament disappears against the bright surface upon which it is superposed. The current used is indicated by the milliammeter, and this can be reduced to terms of temperature; or, s y -7_. .--. -. . ■» ■ ■ " Fig. 137. The Morse Thermo Gage. most usually, the temperature is read directly from the scale. With little practice the eye becomes very sensitive to any difference in the Fig. 138. The Morse Thermo Gage in use. It may also be used in connection with a portable stand. REGULATING TEMPERATURES IN HIGH-SPEED STEEL 169 brightness of filament and the object upon which it is superposed, and temperatures can without difficulty be read with a possible error not exceeding two or three degrees C. With this instrument it i-s possible to prearrange the conditions for the required temperature, to know with certainty when it is reached, and to reproduce the same indefinitely. For the higher temperatures, say above 800 degrees C, absorbent glasses are provided for reducing the dazzling brilliancy of the field. With these accessories the highest temperatures industrially obtainable can be read very accurately. Fery Radiation Pyrometer.— The Fery radiation pyrometer combines some of the features of the optical instruments, with a delicate thermo- couple in the circuit, with a sensitive potential galvanometer whose Fig. 139. Fery radiation pyrometer. Section. E, eyepiece; F, thermo-couples M, concave mirror with aperture; D, diaphragm; T, binding posts; PR, rack and pinion. indications may be read in degrees of temperature. It is virtually a reflecting telescope, the concave mirror focusing the radiant heat of the object under observation upon the " hot " junction of the tiny thermo- couple. The instrument is sighted upon the object and focused by a rack and pinion at the side. There is also a diaphragm for reducing the effective aperture when the instrument is pointed at a very hot object, preventing the overheating of the thermo-couple. Advantages of Optical Pyrometry. — Optical and radiation pyrometers are entirely separate from, and within limits may be at any required distance from, the source of the heat to be gaged; so that they are not at all affected by even the highest temperatures obtainable. Their permanence, therefore, and their reliability, are very great, as compared with most other forms. Distance, as is well known, does of course affect the energy of radiated heat and light waves; but within the limits usually necessitated in the kind of work now under consideration, the 170 HIGH-SPEED STEEL loss is quite inappreciable. Thus it has been shown with a Fery radi- ation pyrometer that the temperature indication of a stream of molten steel was precisely the same whether the instrument was at a distance of three or sixty feet. The comparison lights or filaments, as the case may be, where such are required, naturally deteriorate somewhat with long use; but even then M/ i , c £ » c : 9 a i c ! ■ 1 fi : o > « 5 i > 09 9 00 go 40 j a I > > c *| 20 a 5 in THE CURVE REPRESENTS THE FORMULA r-14S.76|J^ l i0 I t 1 » 1 1 2 1 5 2 2 5 3 3 5 40 Fig. 195. How long a tool should run without re-grinding. This diagram shows that, in order to do the largest amount of work for the lowest all-round cost, a tool should be allowed to cut continuously without grinding at least seven times the time lost in changing the tool, plus the proper portion of the time for redressing, time for grinding, and the time equivalent of the cost of the tool steel. Standard Running Time. — Whatever the form of the tool, it should be possible to establish with a good deal of certainty the standard time for running, at the end of which the tool is to be laid aside for grinding. This time, whether standardized or not, is to be shorter than that required for the ruin of the tool. That is to say, no tool should be run until the edge has broken down, or even begun to break down, but somewhat CONDITIONS OF MAXIMUM EFFECT 231 •saqoui ui 'jooj jo A"poq jo azig rH^OC^OO rH^I t-|QO hH* «HH «H> is O a £ P3 s >> 01 0> o o T) /, 73 a a; rrl w c py a += "o w PS o 01 o o> ^3 -t-5 M 73 _g '+2 pq 01 OJ Ul £ P is 3 ,> T3 73 a rt m cS E3 o o H 3 4) -3 O bC 43 hr X c i-l -»-> u Oj O X at u n Si WJ +-> o H U 73 £ 0) +3 G bfl a Oh >> o K 73 .5 'E a 73 01 J5 43 43 s +2 < M 'an 03 OJ C3 c U CO 43 Sz; o C/J 2 r« 4) +3 to h-1 T3 +-> +J a> P? T3 O +3 W u 03 O a Q O H P m a S s 2 o S o o "3 o u ^S O a> T3 H n ^3 a crt w a C i- rfl £ tt co u 43 a 3 .2 o .5 -+3 m jo C O 232 HIGH-SPEED STEEL short of it. If the matter is left to the judgment of the workman it is necessary to observe carefully the approach of the wear on the lip surface to the cutting edge, and to remove the tool in ample time to avoid ruin- ing the piece of work, or more likely both it and the tool. The length of time a tool will run without re-grinding, it may be added, is not neces- sarily a criterion of its excellence; and indeed it becomes one of the determining elements only when the tool is actually being run at its maximum capacity for speed. Lubrication or Cooling. — In the first years of high-speed steel but little attention was generally paid to the matter of lubrication or cooling. The possibilities of the new tools, as exhibited under ordinary shop con- ditions and without cooling, perhaps were so far in advance of what had been previously accomplished that the possibility of getting still better results was overlooked. There is, however, a considerable gain in cutting speed to be made through a proper cooling of the tool, or rather of the chip at the point where cutting takes place, a gain asserted by those who have given the matter attention, to be as high as 35 or 40 per cent in steel cutting, and from a third to a half as great in cutting cast iron. On many jobs such a possible gain of course is to be taken into account, and provisions made for delivering a cooling agent in suitable quantity and at the proper place. In the case of many jobs, however, such as abound in general manufacturing, especially where the cutting time is brief compared with that during which the tool is not working, no lubrication is necessary; and indeed in most such cases the troublesomeness of a stream of water or oil much outweighs any probable advantage. In all automatic, and perhaps in most semi-automatic machine operations, especially if the pieces be small, the problem is different, and lubrica- tion should by all means be provided, whether the material machined be castings or forgings. When Lubrication Takes Place. — While the word is in common use in this connection, it really is a misnomer to speak of lubrication in con- nection with metal cutting under high-speed conditions, except possibly in such work as milling. Water would have practically no lubricating effect; and it is quite impossible to force oil or other substance between tool and chip in such a way as to do much, if any good, as a lubricant. 1 The purpose of the so-called lubricants in the main is merely to assist in carrying away heat from the place where the work is being done, thus keeping down the temperature of the cutting edge and lip of the tool below the point where softening will begin. In cutting with rotating tools, whose cutting edges are most of the time out of contact with the metal being cut, some actual lubrication evidently takes place, the friction between chip and cutter face being reduced if the exposed 1 Although it is maintained by some, whose experience should entitle their opinions to weight, that lubrication really is effected to a degree worth taking into consideration. CONDITIONS OF MAXIMUM EFFECT 233 portions of the cutting blades are kept completely wet with oil or similar lubricant. At any rate, in milling aluminum a beautiful and clean finish is obtainable, there being no piling up of particles of the soft metal upon the face of the tool, when paraffine oil is used copiously. This is useful also as a lubricant in machining brass. Fig. 196. Correct application of oil or water. Copious Lubrication Necessary. — In the case of milling cutters and the like tools the cooling agent must be supplied in such way, by multiple nozzles or other device, as to keep the face of the cutter well covered, while at the same time it falls upon the chip at the point or line of removal. The latter requirement holds particularly in cutting with lathe and similar tools when a cooling agent is used. There can be no advantage in trying to force a stream under the chip and toward the cutting edge or point of the tool, as shown in the illustration, Fig. 196. The stream must be directed upon the chip just where it is being separated from the body of the piece, and, let it be repeated, in generous amount. The small streams customarily used are quite ineffective, except possibly in the case of very light cutting. It is necessary to deliver gallons of lubricant where it has been customary to deliver pints. The heavy streams serve another useful purpose in cases where the chips come off small or well broken up, in that they carry or float them out of the way. In drilling, for example, it is quite necessary that the stream of lubricant reach the bottom of the hole, not only to cool the lips of the drill, but to float up 234 HIGH-SPEED STEEL the chips. In drilling tenacious material it is better to depend upon a feed sufficiently heavy to allow the chip to come out of the hole in one complete piece, if that be possible. This is especially desirable if the hole be deep. If it does not exceed twice or three times the diameter of the drill, mere flooding in oil will suffice. It is of course necessary in all cases to provide collection pans and suitable drainage. High-Speed Chip not Unique. — In this connection it may be remarked that the chip cut by a high-speed tool differs in no essential respect from that cut by a carbon tool under similar conditions. There was at one time much discussion of this point, based seemingly upon certain superficial differences which arise from the greater amount of metal generally removed and the higher speed at which the work is customarily done. CHAPTER XVI. SPEEDS AND FEEDS, AND RELATED MATTERS. Variables Affecting Efficiency. — The conditions under which metal cutting tools work are so various in different establishments, or, for the matter of that, in the same shop, that generalizations in respect to speeds and feeds are rather difficult, even when jobs are pretty well classified. Certain conditions unquestionably are fundamental; but in the main most of them vary to such an extent that confusion easily results from attempts to apply definite formulas, in working out standards or in applying them to specific operations. Mr. Taylor points out no less than twelve distinct variables affecting the efficiency of chip pro- duction, and indicates their relative importance by the ratios between the higher and the lower limits of speed as affected by each element within the ranges met with in ordinary machine shop practice, as follows: 1. The quality of the metal to be cut — its hardness or other qualities affecting the cutting speed. Proportion is as 1 in the case of semi- hardened steel or chilled iron, to 100 in the case of very soft low carbon steel. 2. Chemical composition of the tool and its treatment. Proportion is as 1 in tools made from tempered carbon steel, to 7 in the best high- speed steel. (This proportion may often be exceeded). 3. Thickness of the shaving, measured by the actual traverse. Pro- portion is as 1 with thickness of T \ inch, to 3^ with a thickness of ^j inch. 4. Shape or contour of the cutting edge of the tool, chiefly because of the influence it has upon the thickness of the chip. Proportion is as 1 in a threading tool, to 6 in a broad-nosed cutting tool. 5. Use or non-use of a cooling or lubricating agent. Proportion is as 1 for a tool running dry, to 1.41 for a tool cooled by a copious stream of water. 6. Depth of cut or the amount by which the piece is to be reduced at the place of taking the chip. Proportion is as 1 with J inch depth of cut, to 1.36 with £ inch depth of cut. 7. Duration of the cut or time through which the tool must last with- out being re-ground. Proportion is as 1 when tool is to be re-ground every 90 minutes, to 1.207 when it is to be ground every 20 minutes. 8. Lip and clearance angles of the tool. Proportion 1 with lip angle 68 degrees, to 1.023 with angle of 61 degrees. 235 236 HIGH-SPEED STEEL 9. Stiffness of tool and work. Proportion 1 with tool chattering, to 1.15 when running smoothly. 10. Diameter of work. 11. Pressure upon the lip of the tool. 12. Possible speed and feed variations in the machine, and its pulling and feed power. Generalizations Necessary. — The very fact, however, of such a multi- plicity of factors makes it essential to arrive, if possible, at some generali- zations, or at least some method of intelligently putting together the variables so as to harmonize them and to work for highest efficiency. The mere recital of experiences, perhaps useful in a way, serves small purpose except possibly to show the limits under a given set of conditions. Depth of Cut i" jl" i" •" I .1.1 ■ T J . 1 . PaTEMTCO Majich B, 1004 DlAMETEl FM POWER. i'lTMih ' |i|i|i|'l i I Speed Co mm FFFVmTkYUITF FEED. onmlfc>i»o/rrfe!><> LATHE No. 43. A-. tU »JU s-U ,-i-, \L, \.i, J-, *U ,-i. yi. I I" Rouw-Notco Tool. &4 5-M i-A-r yo-r »-B-r REV. Of SfWOl* 'A "* ,7lllTilii"ll| ....T..'ilriiM. n ■ I . ■ i . LATHE No. 43. [ 1 I ,1 I' I HIM ill L'l.'l'i.'i.'i-'l pot Curnxa Speed. D»r. WithWatm. i 1 Life of Tool. 2om.ii RC 1- Fig. 197. Barth slide rule embodying the laws deduced by Taylor and his associates. Thus there is little significance in a statement that a certain tool cut soft steel in a lathe at a rate of 200 feet per minute. But if it can be shown that under a given set of conditions susceptible of approximate duplication in a pretty well-defined group of jobs, a tool can be expected to do a certain amount of work, that is to say, can take such and such a cut at such a speed, then something is gained, something established which may be used as a basis for comparison for all jobs falling within the category so defined. Just such a series of laws or formulas Mr. Taylor and his associates succeeded in working out during the course of their investigations, formulas easily applied to the general run of jobs by the aid of a simple slide rule devised for the purpose. Elements Affecting Economies.— The results accomplished through the use of the laws thus established, and the slide rule, have fully justi- fied themselves. And while the work upon which these formulas are especially based is in the main heavy cutting, that is to say, the removal of relatively large quantities of metal from large pieces of stock, they and the tables derived from them are to a very considerable extent available for general use in connection with the ordinary jobs found in general manufacturing; and a systematic use of them is sure to give SPEEDS AND FEEDS, AND RELATED MATTERS 237 surprising results. Reductions in labor cost of a half are not rare; and a considerable reduction in the number of workmen and of machines used is also to be expected in many cases. Such reductions will in large part depend, ordinarily, as well upon a number of other concomitants as upon the cutting speed. These are elsewhere considered, and it is desired here merely to direct attention to them as factors which affect Fig. 198. Barth time slide rule. Used in connection with lathe slide rule (Fig. 197) to determine time required for a given feed, cut and speed. Those especially interested in the use of the slide rules and the formulas developed for use in connection with them are referred to Mr. Taylor's "On the Art of Cutting Metals," and to Mr. Carl G. Barth's "Slide Rules for the Machine Shop as a part of the Taylor System of Management," published in Vol. 25, Trans- actions of the American Society of Mechanical Engineers. the getting out of product and which, if not taken into consideration, may quite nullify any advantage arising from the higher cutting powers of the new tools. Such factors are the size of the piece worked upon, the facility for transporting and storing it within easy reach of the workman, the design of jigs and other holding or guiding devices to facilitate rapidity of handling, the proper grinding of tools, careful inspection of parts, and perhaps others. 238 HIGH-SPEED STEEL Number of Variables Reducible. — The variables affecting the actual cutting, while numerous, can in the average shop be considerably re- duced, and as a matter of fact are actually so in so far as the operations of any particular class are concerned. The depth of cut, for example, in nearly all cast-iron pieces will be about the same for all, unless there should be some unusual condition requiring a different. The standard- ization of tools eliminates the variables resulting from cutting angles and contour of cutting edges except in so far as it may be necessary to use different sizes or special tools. Likewise the composition and treatment of the tool becomes standardized under well-regulated management, and the amount of vibration in machine and tool is reduced to the minimum and thus also standardized, so to speak. The use or non-use of a cooling agent also will be definitely understood. The amount of pressure upon the tool is of so little influence on the cutting speed that it is negligible anyway. So that, outside of the factors which concern the powers of the machine, there I'emain ordinarily but five variants to take into consideration, namely, the quality of the metal being cut, the duration of the cut, the diameter of the piece (generally negligible) in lathe work, the feed, and the speed. In the engineering shop, as dis- tinguished from the general manufacturing shop which duplicates parts in large numbers, of course the variables cannot be so readily standardized. Taylor Standard Speeds. — The standard speeds worked out during the course of the Taylor investigations, given in Appendix F., as already indicated presume long and heavy cutting, conditions under which high- speed tools work at their highest efficiency. But as also pointed out, this class of jobs forms but a small proportion of those found in general manufacturing; and accordingly, for one reason or another, it is much of the time necessary to modify somewhat the standard speeds and cuts established for standard conditions and tools. Thus in a case where the cutting time is very brief relative to that for the whole operation, and the operator has all that he can possibly do to take care of the product anyway, a very high speed is not only unnecessary, but perhaps even undesirable. The gain in such case would probably be mainly through the saving in grindings. On the other hand there will be few cases, probably, where some speeding up is not only possible but desirable. The problem thus becomes so involved that the only safe, way is to accept the standard speeds, and working from these, establish specific ones which shall be particularly suited to the cases in hand. Commercially Practicable Speeds. — What must be aimed at, generally speaking, is the speediest removal of metal commercially practicable — the best speed, feed, and depth of cut, or speed and feed, if the cut is standard — at which the tool will work effectively and with due con- sideration of economical maintenance and the condition of the existing equipment; and all this without especial regard to the total amount SPEEDS AND FEEDS, AND RELATED MATTERS 239 of metal removed. In order to the attainment of such an end it is necessary first to get away entirely from preconceived notions as to cuts and speeds, so as to be governed by the information available through the experiences of others and through data collected in the shop as experience is gained in the applications of the new tools. How Conditions Vary. — The great variations in different shops, not only in respect to equipment but also to the material worked upon, preclude either the tables here given, or the standard cutting speeds from which the Taylor formulas were worked out, being unreservedly accepted. Indeed, they are for the most part entirely too fast for ordinary practice. The speed curves here illustrated, Fig. 199, are also based on experience, 1 I S. ff B 1 Pi 3. ; i e i a f s s ? «S b; o? o i sr S » a »■ S E' r \ P S P 5 P \ \ 7 S? Pi 3. <§ ' 2 g | - p IE ' i I 1 2 5" B s g § H> H « * V le %*% % * K X » V I6 V x H * X H * '/„ V I6* % Ms* Vis Vte*''S2 '42 X V„ A ^f -i t- v A A s * \ L_ y \ V \ \ T \ \ \ J \ A r \ \\ \ \ \ I \ \ \ \ A 4 \ \ i \ \ A j \ \ \ A \ \ L \ i i \ \ \ \ \ \ \ r \ . \ \ \ \ \ T \ \ V 25 50 75 100 125 Cutting Speed in Feet per Minute The Engineering Magazine • Fig. 199. Curves used in one establishment for determining speeds suitable for various feeds and depths of cut, in turning operations. For milling operations add 50 per cent to the appropriate speeds indicated above, taking care to apply the selected speeds to feeds for which they are suited. For boring subtract 10 to 40 per cent, and for drill- ing and reaming, see table of drill speeds and feeds at page 249. The "small diameters" refer to those under 5 inches. and have been in satisfactory use as a basis for the cutting operations in a large factory for a number of years. In another plant the conditions might be quite different, and a different set of standard speeds might have to be worked out. One of the important things in connection with a program of maximum production, therefore, is a careful investigation into the nature of the material used and the establishment of a set of standard speeds suited to those conditions. Speed Meter and Stop Watch. — In the investigations and demonstra- tions connected with the determination of such standards, a stop watch 240 HIGH-SPEED STEEL and a speed meter are absolutely essential. A form of the latter is avail- able by which the speed in feet per minute can be read off directly from the indicator when the contact wheel rolls upon the moving surface, whether cylindrical or plane. The stop watch is best of the kind with two hands, one of which can be stopped independently of the other. These, in the hands of a competent person charged with the determina- tion of operation time for jobs and able to demonstrate unequivocally that the work can be done in the time set, will work many surprises. It is desirable of course also that some way be provided whereby the workman himself can know certainly if he is using the correct speed — that is, in such jobs as are not standardized. In such, the proper combination once given for obtaining the required speed, there should be no occasion for changing so long as the workman is on that particular operation. Nature of Variables in Metal Cutting. — The variables affecting cutting speeds, to be con- sidered in working out such standards, have been already indicated. Mr. Taylor has in his paper considered so fully the influence of each, and the reasons involved, that it does not seem either necessary or desirable here to do more than mention briefly the nature of those influences. Fig. 200. Warner cut meter, for direct reading of speeds. Courtesy of Warner Instru- ment Company. "3 n 1» O i 85 I>0 75 100 1:1' Cutting Speed In Teet per Minute Fig. 201. Curves showing the effect of increasing hardness in "material upon the speed permissible to a tool, for efficient service. 100 is taken as the standard for moderately soft material, and 400 as about the limit of hardness as usually handled in manufacturing. Quality of the Metal. — By the quality of the metal is meant its relative hardness or softness and other characteristics which differentiate cast iron from steel and other metals, and steels of varying degrees of tough- SPEEDS AND FEEDS, AND RELATED MATTERS 241 ness from each other. The harder the material, in general, the slower the possible speed (note the curves in Fig. 201), this latter varying from 10 to 12 feet per minute in the case of chilled iron, to, say, a possible 500 feet in the case of very mild steel. The latter speed, it should be mentioned, is possible only under most favorable conditions and with a very fine feed and cut, and is scarcely practicable, commercially. Under commercial conditions the quality of material varies so much, even in the same shop, and the variations usually are so difficult to detect before the material actually is put under cut in the machine, that little can be done ordinarily in the way of generalization. Castings, for example, which one day come soft and easily machined, another day may come exceedingly hard. Of course in a plant producing its own material, such differences can be reduced to a minimum, if not eliminated. Usually, however, because of them, it is desirable to allow some latitude to the workman in respect to this matter, so that in general it may be necessary to fix a standard speed somewhat lower than could be advan- tageously used regularly if the material came through of a uniform softness. Quality of the Tool. — The quality of the tool involves, of course, its composition, which must be adapted to the special use or be of a good general all-round excellence, and the proper treatment of it in hardening. Evidently a high-grade steel should be selected, and the tools treated uniformly for the same service. This will eliminate all need for consider- ing this as a variable. Use of Cooling Agent. — The use of a cooling agent will permit a con- siderable increase in speeds, as previously stated, ranging from 15 per cent or so in the case of cast iron, to 35 or 40 per cent in that of steel. It is advisable to calculate upon the use of a cooling agent wherever feas- ible — that is, wherever the cost and annoyance would not overbalance the gain. Length of Cutting Run. — In general work the cuts are short, and thus allow the tool to cool off between times. The time during which such TABLE IX. Showing how much a high-speed tool must be slowed down in cutting speed in order to have it last a long time without regrinding. Given the proper cut- ting speed for a cut lasting to find the speed for a cut lasting divide by or multiply by 20 minutes 40 minutes 20 minutes 40 minutes 80 minutes 80 minutes 40 minutes 80 minutes 80 minutes 20 minutes 40 minutes 20 minutes 1.09 1.09 1.19 0.92 0.92 0.84 0.92 0.92 0.84 1.09 1.09 1.19 Adapted from Taylor. 242 HIGH-SPEED STEEL a tool will run without re-grinding, therefore, is likely to be somewhat longer than if it cut continuously. The establishment of a standard time after which tools in general should be ground is difficult except when they are run continuously at maximum speed, and for the most part it will be impossible to lay down hard and fast rules for this factor. In the main it is safe to leave this to the judgment of the workman. Espe- cially if he is working by the piece or for a premium he will quickly learn to note the condition of the cutting edge and to take care that the tool is ground as often as necessary — not grinding the tool himself, be it remembered, but merely setting up a sharp one which he will have at hand for that purpose. Clearance and Cutting Angles. — No important improvement seems to have been made upon the Taylor standard tools (though it is well to bear in mind what is said elsewhere in reference to the Hartness type of tool), so that in the matter of clearance and cutting angles, as well as contour of cutting edge, it is well to undertake few changes, if any, except perhaps in the case of jobs requiring tools of special form. Effect of Chatter. — The possibility of any considerable amount of vibration will reduce the attainable cutting speed by something like 15 per cent. The purpose of course should be to eliminate this item as far as possible through the use of machines suited to the work and not worn beyond hope. In the same connection the possible pulling and feeding power of the machine must be considered. If inadequate, evidently the speed, and more than likely also the feed, will have to be reduced to meet the condition. The use of a steady rest is advised for all work of any considerable length. Depth of Cut Problems. — The depth of cut will, for the most part, be dependent upon the class of work done in the shop, and as pointed out in another place, may need to be increased, a suitable allowance in the size of the piece being made in this case, so as to overcome chatter or riding of the tool. Factors Affecting Feed or Traverse. — Traverse or feed will be governed to some extent by the possibilities inherent in the machine, but a good deal also by the surface required. The limitations are rather narrow, usually, in a given shop. A round-nosed tool, such as will probably be used for most operations, when cutting with a heavy feed leaves a correspondingly rough surface. And since probably most of the work of the kind now under consideration must be done with a single cut, it is necessary to select a feed or traverse which will leave a sufficiently smooth surface. Anywhere from one-sixteenth to one-eighth inch feed will do this, unless the tool be a very small one. Obviously the larger the curve of the cutting edge, the coarser the feed may be in the case of lathe and similar tools. A variation so great as this would seem to make a considerable difference in the possible cutting speed; for this depends SPEEDS AND FEEDS, AND RELATED MATTERS 243 =\^ oer Min.olfcJ 5i^ 150 1 i 130 1 1 1 1 120 110 100 V 1 VI 90 80 ! 1 N s ^ ! 7U 1 00 ' ^ 1 1 i Soft St ikl. 50 40 SO J i I \ i 20 i 1 10 Ha •d Steel Fig. 202. Relation of economical speed to cut and feed. Steel in steel cutting. Fig. 203. Relation of economical speed to cut and feed in cutting cast iron. Both figures adapted from speed sheets made by Samuel Osborn & Co., Ltd. 244 HIGH-SPEED STEEL upon the feed and cut to a very important extent, as already seen. In fact it would, under conditions allowing maximum effects, make a reduc- tion of a fourth to a third in the speed. As a matter of actual practice, however, under the conditions here considered, the speeds will rarely, if ever, be the maximum possible, so that a variation of the magnitude indicated is of no particular consequence — though of course the con- tingency of a possible necessity for changing the speed must not be overlooked. Obviously the power available must also be considered. In milling operations the table traverse or feed has in ordinary practice been ridiculously small, just as it has in the case of drilling and similar operations; though until recently it has been not far below the possi- bilities of the machines in use. The real need for giving each cutting blade enough work to do to keep it in the metal rather than riding over it at times, seems to have been little regarded. It is quite as necessary for the blade of a milling cutter to get a good " hold," so to speak, upon its work, as it is in the case of a lathe or similar single cutter tool. The traverse of course is interdependent with the number of cutting blades and the speed of cutting. About 0.01 inch per tooth per revolution is correct for most work, though with large cutters and plenty of chip clearance considerably more is allowable. Phenomenal Results. — A tool steel vendor reports having obtained a speed of 232 feet per minute, \ inch cut and \ inch feed, on cast iron; a large user of high-speed steels mentions that he has seen a tool running for a full hour, on cast iron, at the rate of 125 feet per minute, removing scale; and another cutting steel at the rate of 270 feet per minute for nearly 30 hours at a stretch. Indeed it has been reported that a tool has cut steel for some time at the rate of 500 feet per minute. Other phenomenal performances also have been mentioned elsewhere. Now all these, and similar performances, are very interesting as showing something of the powers of the new tools. But they are not commercially practicable. In turning castings, as they come ordinaril}' - , 65 to 70 feet per minute is as high a speed, with usual feeds, as can be maintained con- tinuously without too frequent grin dings of the tool. Light cuts may be taken up to 100 feet. Making allowance for variations in hardness and the powers of the machines likely to be used, however, 50 feet is more likely to be satisfactory as a basis to work from. This is easily double what we have been accustomed to under the old conditions. Light finish- ing cuts have been taken with a satisfactory degree.of accuracy as rapidly as 75 to 85 feet per minute, though something rather lower will usually be found better on the whole. At these rates the speeds for soft, medium, and hard irons would be respectively about 90, 50, and 25 feet; and for moderately heavy cuts about 60, 30, and 20 feet. These speeds of course will be modified according to the size of the tool, the precise feed and depth of cut, and the rest of the variable conditions. Working on SPEEDS AND FEEDS, AND RELATED MATTERS 245 pieces of large diameter, for example, the speeds may be increased 10 to 15 per cent. This is not only because in such pieces the direction of the cut changes less rapidly, but because even in sand molds small pieces have a tendency to chill more or less, and the cut in work of this sort also being customarily lighter than on heavy pieces, the tool is actually working on iron averaging a good bit harder than if cutting on a large piece. Turning Chilled Iron. — Chilled iron can be turned at anywhere between 3 and 10 or 12 feet per minute. The rate of cutting is of small conse- quence, since most of the time is used in changing the tool anyway. A tool that will hold up at 3 feet will be almost certain to do quite as well at three or four times that speed. It should be remarked in passing that the turning of chilled rolls is accomplished by the use of tools of special design, ordinary tools being of little use generally. The common method is to use a square section of steel for the tool, grinding the longi- tudinal edges to sharp right angles for cutting edges, and rigidly wedging this unique cutter against the surface of the roll. The tool is not trav- ersed, but its position is changed each time it has finished a section of the roll. For cutting grooves and spirals, the end of the section instead of the longitudinal edge is forced into the iron. Speeds in Steel Turning. — In steel cutting, speeds have been attained very much higher than on cast iron. The reasons for this have been already assigned. On mild steel light cuts at 80 or 90 feet are not infrequently taken, and 100 feet or even more are sometimes used. Taylor's tables allow 128 feet per minute on soft steel with a f inch tool, cut | X | inch. These are not usually practicable commercially except under conditions allowing maximum effect. Roughing cuts, say about f X s 3 5 inch, with a 1-inch tool, can be taken at a maximum of about 100 feet per minute on mild steel, while 70 to 75 feet is considered good practice. Medium steel, say 3 to 5 point carbon, under customary conditions, can be cut at approximately 70 and 50 feet for light and heavy cuts respectively, though rather less is likely to be more satis- factory; while hard steel, say 7 point carbon and above, is not usually cut faster than about 35 feet in light cuts (say about | X i inch) and 25 feet in roughing. Cutting Malleable Castings. — In malleable castings 80 feet is about all that can be expected on small pieces, where light cuts are taken. Heavy cuts, as already shown, are to be avoided in castings of this kind. On large pieces the speed may under favorable conditions be made as high as 90 or even 100 feet per minute. If a cut as deep as \ inch be taken for any reason, the speed will be nearly half that indicated. Brass, Aluminum, and Alloy Steels. — Speeds for brass and other soft materials will be somewhat greater than those suitable in cutting steel. The alloy steels, like chrome and nickel steel, now coming into large use 246 HIGH-SPEED STEEL for machinery parts, vary so greatly in characteristics that it is quite impossible to lay down at present any general statement as to permissible cutting speeds. Aluminum can be worked at about four times the speed and double the feed possible with mild steel. Fiq. 204. Use of multiple tools in turning cast-iron cone pulleys. Speed on largest step t 120 feet per minute. Use of Multiple Tools. — The use of double or multiple tools allows a distinct gain in speed in turning operations. The interrelation of speed and feed makes it necessary to reduce the former when the latter is Fig. 205. Lathe fitted for rapid roughing and rapid finishing. increased. A single tool taking a chip f inch wide (traverse f inch per revolution) must run considerably slower than two tools, each feeding at fV inch per revolution. Doubtless there is some difference in the amount of power absorbed; but this is a matter of relative insignificance. The advantage of using multiple tools in long cuts has never been recog- SPEEDS AND FEEDS, AND RELATED MATTERS 247 nized as it should be. The advantage under the new conditions should be evident enough to warrant a large use of them, especially in operations involving the removal of large quantities of metal. Reduction of Speeds in Certain Operations. — It has doubtless been observed that the discussion of speeds and feeds thus far has been con- fined for the most part to turning operations. With the possible excep- tion of milling, where the cutting edges work intermittently and are for Fig. 206. Multiple tool carriage, as applied to "Lo-Swing" lathe. See Fig. 207 for data as to this particular job. the greater part of the time exposed to the cooling effect of the air or a cooling agent, these permit greater speed than any others in metal cutting. It is therefore necessary, in setting standard speeds for other kinds of work, to make allowance for the variations in conditions — ■ variations not provided for in most tables of cutting speeds now available. Inter- nal cutting with a boring bar, for instance, must be considerably slower than external turning of the same diameter. In outside cutting the tool may have ample clearance, or it may even ride upon the finished flank — a condition impossible in the case of internal cutting. Further- more, in order to secure a sufficient clearance it is necessary to give the cutting edge a more acute angle, or else to give it less rake than that of the corresponding standard tool. Either condition involves a slower cutting speed, the reduction being something like 10 to 15 per cent if the bar is exceedingly rigid. If not, the loss is more likely to be nearer 40 per cent. 248 HIGHSPEED STEEL As to Drilling. — Something of the same sort is true of rose reamers, and of twist drills also, to some extent. In drilling, little attention need be paid to this point, however, since but few machines are now in use which -21H K — 3--X2}£>K ~ JL Pinion Shaft -\VA- ->k-2>+ -f~] £*P LET WIDTH OF TOOL=A" J AND RADIUS OF POINT=R THEN u FOR BLUNT TOOL R^A-JL FOR SHARP TOOL R=\ A-% For cutting hard steel and cast iron, these tools are ground to the following an- gles : Clearance angle 6°, back slope 8°, side slope 14°. For cutting medium steel and soft steel, these tools are ground to the following angles: Clearance angle 6°, back slope 8°, side slope 22°. Fig. 212. Outline of cutting edge, Taylor standard round-nose tool. Standard Angles for Cutting Tools. — The matter of lip angle (for definitions of terms used in connection with lathe tools see Fig. 213), for instance, while not of the degree of importance sometimes attached to it, has some influence upon the permissible cutting speed and the excellence of the work done, and a very, considerable influence upon the 256 HIGH-SPEED STEEL lasting quality of the cutting edge and the stresses imposed upon tool and machine. The clearance angle, from the heed for the greatest possible support of the cutting edge, must be small, say about 6 degrees, as in the Taylor standard tools; while not yet so small as to prevent the I HE6.1- OF TOOl Z2TEARW* / Fig. 213. Key to terms used in connection with lathe tools. From Taylor, i 8 r "-i. HARD STEEL AND CAST IRON SOFT STEEL Fig. 214. Angles for cutting cast iron and hard steel compared with those for work in soft steel. tool being readily fed into the work and to cause the flank of the tool to rub too much against the finished work. The lip angle of a tool will depend in a considerable degree upon the work it is to do. A small angle reduces the pressure upon the tool and perhaps allows some in- crease in speed; but other considerations, chiefly the necessity for having FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 257 a sufficiency of metal at the cutting edge to carry away the heat generated, and at the same time stand up to the work without crumbling, indicate a rather obtuse angle. The harder the metal being cut, the more blunt the cutting angle should be, the general rule being that the angle should be just as sharp as it may be without crumbling or spalling so rapidly as to impair the efficiency of the tool. The angles recommended by Taylor for the several typical classes of work ordinarily found in a shop, are given as follows: Cast iron and harder steels (say 0.45 carbon and up), lip angle 68 degrees; clearance 6 degrees; back slope 8 degrees; and side slope 14 degrees. Softer steels, lip angle 61 degrees; clearance angle 6 degrees; back slope 8 degrees; side slope 22 degrees. For cutting chilled iron, 86 to 90 degrees lip angle. Tire steel, and similar hard steels, lip angle 74 degrees; clearance angle 6 degrees; back slope 5 degrees; side slope 9 degrees. Extremely soft steels (say 0.10 to 0.15 carbon), lip angle about 61 degrees, and perhaps less. Relation of Side and Back Slope. — It is to be noted that the side slope recommended in these tools is considerably greater than the back slope, and varies also considerably from the angles heretofore customarily employed. The relatively steep side slope allows the tool to be much more frequently ground without weakening it, allows the chip to slide off in a line avoiding the tool post or holder, helps to correct the tend- ency of the tool to side deflection by throwing the pressure line within the base of the tool, and reduces the feed pressure. The greater blunt- ness in tools for cutting soft cast iron, compared with those used in cutting soft steel, is also noteworthy. 1 Hartness Type of Lathe Tool. — Excellent results are obtained in many shops through the use of lathe tools of an entirely different type from those just described, cutting the metal in a way essentially different. In these tools, designed, it seems, by Mr. James Hartness in connection with his Lo-Swing lathe, there is no clearance, no forging, and the cutting angles may be of almost any degree of acuteness, in most cases much smaller than is customary. The cut is taken straight sidewise, the tool feeding along the periphery of the rotating work, and slicing off a band of a depth and thickness corresponding to the feed and depth of cut. 1 Space does not permit here a discussion of the various experiments and experi- ences furnishing the reasons for the standards (those recommended and used by Taylor as standard cutting tools) given above. Any one caring to pursue the subject further will find much of interest in Mr. Taylor's address or report "On the Art of Cutting Metals," already mentioned.. Another paper of very great interest, by Mr. James Hartness, considering the nature, cutting angles, and utility of the type of tools re- ferred to in the following paragraph, was read before the same society at the 1908 meeting. 258 HIGH-SPEED STEEL The ribbon chip is removed at the front (that is, by the edge of the tool) by real cutting, the nose of the tool being forced into the metal and wedging it away from the main mass; while at the finished surface of the work the action is comparable to that of shearing, leaving, however, a Fig. 215. Examples of Hartness tools, with sharp cutting angles. satisfactory finish surface. The end or nose of the tool rides against the flank of the finished piece, and by giving the cutting edge and lip a suit- able positive or negative back slope, the pressure against the tool, tending to force it out of the work, can be minimized; at the same time Fig. 216. Characteristic chips. Those at the left were made by a diamond point tool having 70 degrees cutting angle. Chips at the right made by a no-clearance tool, 45 degrees cutting angle. From "Ma- chine Building for Profit," by James Hartness. the cutting angle can be so selected that the tool will actually feed itself into the work, entirely eliminating the feed pressure, the feed drive being required mainly in starting the cut and in maintaining it uniform. By mounting the tool in a holder or post in such a way as to afford some freedom of movement, lateral vibration in cutting is almost if not en- FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 259 tirely eliminated, and the cutting edge of the tool therefore is unimpaired by reason of chattering, which is relatively a much more important factor in its destruction than heat is, in tools not employed in exceed- ingly high speed or heavy feed cutting. By suitably selecting the cutting angles the tools also become, to a certain extent, self-sharpening, the lip and flank riding against the cut surface wearing along with the edge itself, though perhaps less rapidly. For many kinds of work such tools are run with faces nearly straight, on the principle of the side tool. In other cases it is found desirable to give the tool a slight round. In turret lathe practice, when working on bar stock properly supported by back rest, the corner of the tool is left nearly or quite square. Utility of Hartness Tools. — Tools of this type are inexpensive to make and seem to have a rather wide range of usefulness in the ordinary metal-working shop in connection with appropriate forms of turret lathes doing repetitive work of moderate size and not of the kind where the so-called rapid reduction, the speedy removal of relatively large quantities of metal, is necessary or economical. In diameters under, say 3^ inches, work up to 3 feet in length can be advantageously turned with such tools, while in diameters ranging from that given up to, say 20 inches, the lengths conveniently turned are up to nearly a foot. Diam- eters smaller than one inch, or possibly a little less, are riot so well adapted to the use of these tools. 1 The Problem. — Taylor Standard Tools. — The problem in designing lathe tools, and those of similar use, is in the main one of so harmonizing the various elements affecting the form, the shape of the cutting edge, and the lip angle, that the highest all-round efficiency shall be obtained with a minimum number of different tool forms. The tests and expe- riences of Mr. Taylor and his associates, and of many others who have adopted his standard shapes, leaves no doubt as to their all-round efficiency in roughing and rapid reduction work. They require con- 1 The experiences and observations of Mr. Hartness have resulted in his formulating the following conclusions with respect to the advantages of this non-clearance type of tool: Relieves the edge from one-sided pressure. Prolongs the life of the cutter by allowing abrasion on its face without producing negative clearance. Converts the lip angle into a cutting angle, which for a tool of given form consti- tutes a gain of from 5 to 10 degrees in cutting angle. Extends the range of the side tool (a tool of this type is really a side tool), which gives the minimum stress. Makes possible the use of acute-angled tools, thereby increasing the output of machines which have been limited by lack of pulling power. The reduction of the cutting and separating stresses increases the accuracy (or output, which is generally interconvertible with accuracy) on nearly all lathe work. This reduction of stresses also increases the output, which has been limited mostly by the frailness or slenderness of the work. 260 HIGH-SPEED STEEL Fig. 217. Type of lathe tool much used on the continent, but not equal in efficiency to tools requiring more forging and permitting a greater number of grindings. Fig. 218. Detailed dimensions of Taylor standard one-inch round-nosed roughing tool. Forged outline shown by dotted lines. FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 261 siderably more forging than tools customarily used; but this very thing is the result of a compromise in design whereby many grindings with few dressings are possible before re-fettling and re-hardening become necessary. 1 Of course in the ordinary run of shops, where the variety of work is large, there will inevitably be many jobs which can be better done with tools of other or of special design. It is well, however, to bear in mind always that a multiplicity of tool forms greatly complicates the tool problem; and that when all things have been taken into account it may after all be quite as economical to use a standard as a special tool. Tool-Holder Stock. — In the early days of high-speed steel the tendency was to use it very sparingly, and for the most part in the form of tool- holder stock. This method permitted the use of a minimum of stock (and a consequent minimum expenditure, as it was thought) and neces- sitated little expense in tool making, since the stock was usually bought unannealed and was put to work after nothing more than grinding to shape. The lightness of the stock which could be used was a manifest shortcoming, which was soon remedied by the design of other tool Fia. 219. A patented tool holder which allows close contact of tool and holder and insures an unusual degree of rigidity in the tool. holders large enough and heavy enough to meet the requirements in this respect. The use of the unannealed stock, without treatment, has been found to be unsatisfactory; and the usual practice now is to treat tool-holder stock precisely as other tools are treated except that little forging, if any, is attempted. For many uses, especially for light cut- ting, the use of such stock, in suitably designed holders, is permissible and gives good results, though usually not so good as can be otherwise obtained. 1 Following are the affirmative considerations set forth by Taylor in connection with lathe tool design: The bar from which the tool is forged should be one and one-half times as deep as it is wide. The cutting edge and the nose should be set well over to one side in order to avoid the tendency under pressure to upset in the tool post. That shape should be given preference with which the largest amount of work can be done at the smallest combined cost of forging and grinding. Forging is much more expensive than grinding, therefore a tool should be designed so that it can be ground the greatest number of times with a single dressing and the smallest cost per grinding. The best method of dressing a tool is to turn its end up high above the body. 262 HIGH-SPEED STEEL Defects of Tool Holders. — The stock of a lathe (or similar) tool not only serves to support the cutting portion, but also to conduct away a considerable part of the heat generated in cutting. In order to do the first effectively the stock must, as has been pointed out, be strong enough to prevent springing and chatter. Holders for this reason, if used at all, should be made of chrome or other very tough steel. Also, if used, they ought to be so designed that the tool itself will fit snugly, Fig. 220. A type of composite tool, suited to scraping. or even tightly, into the slot provided, so that there will be close con- tact on all sides between tool and holder in order to allow the heat to be conducted freely from the one to the other. Under the very best conditions, in this respect, there is likely to be something still wanting; and this lack is emphasized by the fact that in all such -mechanical com- binations of tool and holder there is introduced one additional joint in the circle which includes tool, tool post and slides, lathe bed, head stock, spindle, chuck or dogs, and the work itself, and thus brings in an addi- tional element of negative force in high-speed cutting where especially rigidity is essential. FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 263 Compound or Welded Tools. — The better practice seems to be to use a stock of chrome, nickel, or other tough steel with a nose or cutting end of high-speed steel either electrically or autogenously welded on (see Figs. 151 and 152), or possibly joined by some other method of rendering the parts homogeneous, as described elsewhere. Tools so made can be forged to standard or special shapes and re-fettled as necessary. They have the requisite strength, avoid introducing additional holding devices with the consequent liability to movement, allow the heat to be conducted away from the cutting edge without interruption, and are in practically every respect equal to solid high-speed steel tools. The cost probably is a little higher than that of tools used in holders, but the advantages Fig. 221. Types of milling cutters best made from the solid stock. clearly outweigh any slight difference in this respect. Tools with cutting surfaces brazed to ordinary steel bodies or stocks have been in use for a number of years in planer and similar work (see Fig. 149), and to some extent also in turning operations. Where no forging, or but a .slight amount, is required, these are very satisfactory. The method does not commend itself in connection with tools of the Taylor standard shapes, or others except those of simple form. Composite Rotary Tools. — In milling cutters and other rotating tools of large diameters there is opportunity for economizing in the use of high-speed steel by the use of this material for the cutter parts only, the body of the tool being in a large proportion of the cases quite as well made of cheaper material. The smaller cutters, say those below 4 or 5 inches diameter, and most irregular shapes, are usually made solid for the obvious reason that they cannot be conveniently or economically made with inserted blades, especially if held by mechanical means. A certain amount of room is required, which in small diameter tools is not available even if the number of cutters be reduced considerably below the customary; and the necessary smallness of the securing parts usually 264 HIGH-SPEED STEEL tends to insecurity in the holding device. The objection to tool-holder lathe tools touching the lack of heat conductivity, does not hold in respect of milling and similar cutters. The cutting edges are at work for a Fig. 222. Something unusual in the way of large milling cutters, 9x2 inches. Made from the solid because composite cutters would not stand up to the work required. The peculiar form of the teeth pre- vents dragging and gives free cutting angles all round. Fig. 223. An interesting composite milling cutter. So designed that the blades are easily removed and ground simultaneously by the aid of a grinding fixture. Courtesy of Mr. William G. Thumm. relatively short time during each revolution, and are exposed for the remainder of the time to the cooling action of the air. On this account they do not tend to become heated during a long run. In all the larger sizes, therefore, where it is convenient and feasible to secure the cutting blades mechanically, this is usually the preferable method. This the more because of the difficulties inherent in hardening not only these, FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 265 but most other large tools of intricate form. All such are to a greater or less extent susceptible to cracking, which though it may be reduced to a minimum by use of the barium hardening method or extreme care with other methods, still may possibly occur, the cracks, when they do occur, often not becoming manifest until the tool has been put at work. Fig. 224. Type of inserted-blade milling cutter. This composite method permits the use of a tool body, with properly made recesses for blades and means for securing them, once made, to be used with an indefinite number of cutter sets, in this. way reducing the actual tool cost on any given job very materially below what it would be if the ordinary solid carbon cutter were used. Fig. 225. A cold saw (Taylor-Newbold) with inserted teeth held in place by soft metal. Methods of Securing Cutter Blades. — The methods of securing cutter blades are not only divers, but of various degrees of excellence. Brazed cutters have been used, and in a few instances also welded ones. These 266 HIGH-SPEED STEEL methods produce the strongest tools; but they have the disadvantage of making the body, with its expensively cut recesses, usable but the once. Holding the blades in place by means of screws is, in general, to be avoided because of the well-known tendency of threaded parts to work loose and sometimes to require replacing with new and over-sized ones. Screws in some cases are used for the purpose of expanding the peripheral sections of a cutter in such a way as to grip the blades between ; and this is an excellent way. Expansion dowels are similarly used, as Fig. 226. Tooth of Taylor-Newbold saw broken. The tool or holder will break before the soft metal bedding the tool in the holder is materially damaged. also are wedges. In the latter case, however, there is more or less diffi-- culty in extracting the blades when worn out — a matter of some impor- tance. For many uses it is sufficient to make the blades of such form and size that they are merely pressed into place with a forced fit. The method of imbedding the cutter blades in soft metal, hereafter described, also has advantages. Helical Milling Cutters. — For side facing mills and others needing but a narrow peripheral cutting face, where the blades are of little length, they are sometimes straight, but are most often set at an angle so as to secure the advantage of a partially shearing cut. The same thing has been tried with long cutters, but in this case it is impossible to make a satisfactory mill without the use of helical blades. The slope and lip angle of a mill- ing cutter blade obviously should be uniform, or nearly so, throughout its length. In very short straight blades this will be the case nearly enough for practical purposes, but where the cutters are long, unless bent to a helical form, if set with the front face at an angle with the axis, the slope and angle are uniform at. no two points along the cutting edge but are modified to such an extent that the length of blades so set is thereby very FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 267 limited, and mills so made are liable to excessive vibration and chatter. In the case of helical cutting edges the front slope and lip angle can be maintained uniform throughout their length. Heretofore, however, it has been rather difficult to manufacture a cutter with blades of this form satis- factorily and effectively secured. This is now done by milling (or sometimes planing) key or pocket shaped helical slots, relatively large as compared with the blades, and anchoring the blades to the holder or tool body by filling the vacant space with some soft alloy like type metal and compressing it to insure a perfect imbedment of the blades. Anomalous as it may seem, tests have proved that blades, or Fig. 227. Development of front slope from to a positive front slope in the case of a straight inserted cutter blade set at an angle to the axial plane. Courtesy Tabor Mfg. Company. Fig. 228. How the front slope varies from the maximum at R 1 to the minimum at R* in a straight blade set at an angle of 20 degrees to an axial plane, while it remains constant throughout the length of the helically curved blade set at the same angle. The condition arising in the former case limits the possible length of the blade. Courtesy of Tabor Mfg. Company. even holder, will usually break before the anchorage is materially dis- turbed by stresses upon the cutter blades. Blades so held are removed without trouble when in need of replacing. A well-known cold saw, it may be said in passing, has its teeth secured by a similar method. Renewing Inserted Cutters. — This matter of renewals is dependent very largely upon the relative amount of cutter blade which can be ground away, and the method of grinding — that is, whether face or back grind- ing. The latter is by far the better, a very small amount of metal re- moved serving to sharpen the tool where a heavy face grinding would be required to accomplish the same result. The life of a set of cutter blades of this sort, then, depends upon the number of grindings that can be given it, which is to say, upon the overhang or distance the blades 268 HIGH-SPEED STEEL project from the housing. This distance will naturally be governed by the work to be done, that is, by the stresses to be provided against. Under ordinary conditions a projection of one and one-half times the thickness of the blade is not too much. Cutters not helical in form may not permit quite so much projection. ^-CLEARANCE LIP ANGLE Fig. 229. Unique method of holding inserted cutter blades in position by imbedment in soft metal. The front slope is given to the blades by curving the face. Courtesy Tabor Mfg. Company. Maximum Number of Blades. — Where metal is removed as rapidly as it should be with milling cutters of the high-speed type, the question arises as to clearance for the chips as they are cut away. The pitch ordinarily used in carbon steel milling cutters is insufficient, especially in cutting soft metals. Furthermore, the mechanical fastening of the blades also limits their number. Just what should be the number of cutting edges does not seem to have yet been scientifically determined. This much is known, however, that the best results are not obtained where the number of cutters is large. Coarse feeds are coming more and more to be the rule in all except fine finish milling; and for such work evidently the number of blades will be governed to a large extent (in connection with the considerations already mentioned) by the load it will be practicable for each blade to carry; for the abutments for the several blades, taken together with the thickness of the blades them- FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 269 selves, must be sufficient to withstand the probable stresses imposed. For most classes of such cutting operations the number Of blades will vary from about one for each inch on the periphery in small diameters, say 4-inch, to about one for each 1£ inch of periphery when the diameter is as great as 10 inches. This would give about 14 blades for a 4-inch nominal diameter, 16 for a 6-inch, 18 for an 8-inch, and 20 for a 10-inch cutter. The arbor hole should never exceed one-half the nominal diameter. In the case of side and face mills, where the chips are not confined, a greater number of cutters is allowable if considera- tions of strength and proper securing permit. Evidently a solid cutter can have a greater number of teeth than one with inserted blades. Such an increased number, however, is not at all necessary, for the work done with one of coarser pitch will be quite as good as that with the finer. Furthermore, the cost is in the neighborhood of a third less. Usually also the life of a coarse pitched cutter will be considerably, not infrequently several times, greater than that of one with the greater number of cutting edges. Rake for Soft Metal Cutting.— Milling cut- ters (and most other tools as well) intended to work on aluminum, and perhaps on other soft Fig. 230. A Tabor inserted blade milling cutter of some size. Scarcely feasible to make it solid. Blades secured by im- bedment in type metal. metals also, require more rake than those cut- ting iron and steel, in which latter case 5 de- grees is about right. An angle of 45 degrees from the vertical gives a beautiful clean finish when paraffine is used as a lubricant. The removed material does not under these con- ditions pile up on the face of the cutter, rough- ing its surface and preventing the cutting of a clean chip as often happens otherwise. The Fig. 231. High-speed milling cutters . ,. ,. , ,, , , T should have about 5 degrees of pitch or distance apart ot the cutters also IS slope to the cutting lips. ' .. . ,, . , , , a • -\ . necessarily greater, three teeth to a 4-incn cut- ter being amply sufficient, while a cutter as large as 10 inches requires but 6 teeth. More than this number is unnecessary to secure a good finish, and would be in the way of chips, preventing their clearing out properly. 270 HIGH-SPEED STEEL Nicking the Cutting Edges. — The question as to whether the cutting edges of long mills should be nicked or not is still open, though it would seem that if the cutters were designed with a front slope such as. to bring down the chip pressure there would be no occasion for breaking up the chip. This front slope is an important factor in the efficiency of milling cutters. Interlocking Mills. — Wide face cutters, when solid, are preferably made in interlocking sections, as also are those used for producing Fig. 232. The gang milling cutter, built in sections in order to simplify problems of manufacture, hardening, grinding and maintenance. finished surfaces with a variety of curves, slopes, or angles. In the latter case gangs are used, a separate cutter for each surface or curve, usually. The difficulty of properly hardening a single cutter of extreme length or intricate form would be sufficient reason. A further one is that in case of damage to a portion of such a cutter, if made in sections the damaged part can be replaced without the expense involved in the making of a complete new cutter. The Case of Rose Reamers. — Rose and similar reamers have several points in common with milling cutters, yet in certain respects are in a class by themselves. Those used in a vertical position, for enlarging and truing cored holes, discharge their cuttings freely; but those working FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 271 in a horizontal position often present a difficulty in this respect, and require greater clearance in order to avoid clogging and other troubles. As in milling cutters, in order to secure a sufficient clearance, and at the same time sufficient strength in the backing or abutments supporting the cutting edges, the number of the latter is reduced as much as a third, frequently, below the customary. Composite Reamers. — Small sizes, say up to H inch, are preferably made solid, and above that diameter with inserted blades, though some as small as | inch are made composite — and adjustable at that. The manner of insertion may be either mechanical or intimate, as with milling cutters, Fig. 233. Matthews expanding shell core trill or reamer. Expansibility secured through slotted shell in combination with an expansion bolt. though in the former case the mode of holding is essentially different. If intimately attached, so as to make a practically homogeneous tool, either by brazing or by welding, the effect is that of a solid tool which may be ground off without reference to saving the core or body. This is convenient, especially when the operator is required to grind his own tools — as he should not be. By the use of suitable collars and other well-known devices the blades may be secured at both ends, and when ground away until no longer serviceable are quickly replaced by another set. Likewise there are reamers whose blades are forced with a drive fit into the grooves cut in the body of the tool; and still others with blades wedged in or secured by screw devices. Almost any of these are efficient in light or moderately heavy work, particularly if used as floating reamers for sizing rather than for boring. When, however, extra heavy service is required there is some difficulty in securing the blades mechanically so that they will stand up to the work. It is desirable in this case that the blades be brazed or welded — and to cores strong enough not to twist off at the shank. Material for Stem or Body. — To afford the requisite strength in tools of the last mentioned kind, the body and shank are best made of a 272 HIGH-SPEED STEEL tough chrome or similar steel. Machinery, and even tool steel shanks, not infrequently twist off under the heavy stresses imposed by severe duty. For ordinary work under customary conditions the bodies are strong enough if made of machinery, or at best of tool steel. Occasion- ally it may be desirable to make them in the form of steel castings, though usually the ordinary method will be followed. Bronze metal bodies, with the slots for the blades milled in the customary manner, also are in successful use. The blades in this case are brazed into the recesses. Cast or malleable iron bodies are to be avoided. Clearance and Relief in Reamers. — High-speed reamers especially, and perhaps more than the slower carbon steel sort, require a sufficient and proper clearance or relief. Too much will result in chatter, while too little will lead to binding in the hole and consequently to short life. Flat relief, that is, a flat land behind the cutting edge, or front face of the flute, is not nearly so good as a curvilinear, the so-called eccentric or radical, which not only gives better support to the face of the flute but also helps to steady the tool and produces a better hole. Expansion or Adjustable Reamers. — The work of a rose reamer is essentially different from that of either a milling cutter or of a drill, since the former removes but a relatively thin skin of material from the Fig. 234. Smith adjustable reamer assembled, and parts of same. inside of a hole already existing; and unlike either, it takes this off not with a broad cutting edge, but with a small corner of the cutting lip at the periphery of the tool. In consequence nearly the whole wear comes just at that point in the reamer which gives the size to the hole. And since the only way in which work of this character can be kept within FUNDAMENTAL CONSIDERATIONS IN THE DESIGN OF TOOLS 273 the requisite limits of precision is to keep the tool also within those limits, it is necessary to keep such reamers well sharpened by frequent grind- ings, or to provide for taking up this wear by some method of expansion which will again give the required diameter. In most cases where the nature of the work is such as to permit the use of tools of that type, the expansion reamers are preferred. The simplest form of such a tool, perhaps, is that in which the blades are welded or brazed to a slotted shell provided with a taper plug and take-up screw. In. another well- known tool, with removable blades, the expansion is accomplished by rotating a locking cam bolt with cams corresponding to the blades. Twist Drills. — In fluted twist drills there has been little change from the standard form previously in use, though it is desirable that the lead of the flutes be given some advance over the customary one of about seven times the diameter. The smoothness of finish somewhat affects the possible speed and therefore the efficiency of drills. Provision should therefore Fig. 235. Chard spindle drill. be made for a finish approximating a polish. The one especial feature wherein the high-speed drill can well differ from the ordinary is in the thickness of the web. In order to give the greatest possible degree of stiffness and strength where most required, it is a desirable, but by no means an universal practice, to increase the thickness of the web grad- ually toward the shank. It is important that the web, especially when the drill has become somewhat shortened, should in grinding be thinned at the point in order to minimize the tendency to split which sometimes manifests itself at the high pressures required to feed these tools into the work. Likewise it is not only important, but essential, to provide for the grinding of a proper clearance to meet the requirements of the feed intended to be used, ana, as explained elsewhere, for so varying the clearance angle from center to periphery as to allow the drill to feed properly into the work at all points along the cutting edge. Twisted Drills. — For a time at least one maker of high-speed steel rolled sections of such shape that when suitably twisted and provided with a shank, a fluted twist drill was produced which required only to be finished and ground, the cost being surprisingly small. This was, in a sense, a reversion to the original method of making drills of this type. The maker after a time ceased rolling the drill stock section; but 274 HIGH-SPEED STEEL others have since then taken up the idea and now several forms of drills are manufactured in this way, ranging from those twisted from flat stock to others rather closely approximating the standard milled fluted drill. There is no difficulty in producing a shank suited to any require- ments, from the regular taper to special forms adapted to use in con- nection with special collets or chucks. In one style the lead of the twist is considerably increased at the shank end so as to form a good bearing, which when ground not only fits the standard taper socket, but Fig. 236. Twisted drill made from a flat bar of stock throughout its entire length. Shank ground to standard taper. by reason of the tendency to untwist under the stress of work, it actually grips the taper seat more firmly than a solid shank drill does. Other styles have hot pressed or slightly forged shanks, flat or flat-taper, as the case may be. Economy in Twisted Drills. — The manufacture of drills of this type obviously is a much simpler and cheaper matter than that of the ordinal fluted drills. The amount of metal required is only about one-third as much by weight, which of itself is a matter of consequence; and the twenty odd operations involved in the manufacture of ordinary drills is reduced to a small fraction of that number. No expensive or elaborate special outfit is required to make them, and it is therefore possible to produce drills of this type in many shops otherwise not equipped for drill making. In addition to all these things, it would appear that drills thus twisted are stronger than those milled from the solid cylin- drical stock and that tangs very rarely twist off or stems break under the strains of their work. Obviously the grinding and finishing must be as carefully done as in the case of the ordinary type. As to Other Tools. — There does not seem to be any reason for depar- ture from traditional lines in the design of wood working and of metal working tools other than those used for rapid cutting, except in so far as it is desirable to make all these tools so far as permissible on the com- posite or built-up plan. Small dies, punches, shears, and the like tools are preferably solid; but large sizes are just as well, or better, built up in such a way that the parts subjected to wear may be renewed from time to time as required. CHAPTER XVIII. THE NEW MACHINE REQUIREMENTS. Limited Use of High-Speed Tools.— The revolution in machine shop practice, so enthusiastically predicted for some time, unquestionably is arriving; but, it seems, rather more slowly than might have been expected; and it is as yet manifest in spots only, so to speak. Considering the very great advantages obtainable by the extensive and intelligent use of high-speed steel tools, it is surprising, not to say disappointing, from the efficiency point of view, that they are as yet used so little and so ineffectively in general manufacturing. In such plants as have for their principal product heavy forgings or machinery, or other products Fig. 237. An extraordinarily large and heavy lathe in its day. Built about 1856 by the Phcenix Iron Works, Hartford. Compare the weight and build of this machine with a modern high-speed lathe of about the same swing, as seen in Fig. 238. comparable to them, the new steels have a large and not infrequently exclusive place. In a relatively few shops turning out products of a different character, say such as would be typified by agricultural ma- chinery, sewing machines, and watches, advantage is also taken of the new steels and their high efficiency. In the average factory, however, the extent to which they are used is astonishingly small. There are indeed places where high-speed steel seems never to have been heard of, or where the management is so ultraconservative as apparently to deserve being under suspicion of inefficiency and to need reminding of the adage which says something about penny-wise and pound-foolish. 275 276 HIGH-SPEED STEEL There are of course many considerations affecting the use of high- speed steels, and these are discussed in another plage. One of the most important relates to the nature and effectiveness of the machine tool equipment; and it is this which particularly concerns us at present. Incapacity of Old Equipment. — It scarcely needs mentioning that ma- chine tools designed under the old regime, to meet the requirements and reach up to the limitations of the old tools, are utterly incapable of using efficiently the new tools, with their doubled, trebled, and even quadrupled powers. It may be necessary or expedient to use such equipment even in connection with high-speed tools, as pointed out in another place; Fig. 238. Example of powerful and massive lathe especially adapted to use high-speed steel tools. Double head, one of which has traverse. Feed and headstock traverse both secured through an aux- iliary motor seen at the farther end of the machine. 80-inch Niles driving wheel lathe. but in the acquisition of new equipment there certainly should be no hesitation in selecting that which will measure up to the full powers of the new tools, and then to use those tools at their maximum efficiency. Sufficiency of the New Types. — During the first years of high-speed steels, machine tools fulfilling the new requirements were not often to be found. Builders were cautious in the matter of new design and the expense of putting out new types of machines — though it may be sup- posed that they were no more conservative in this respect than the market demands made prudent. The early attempts at adaptation to the new standards were mainly in the way of modifications of existing designs. There was some increase perhaps in the weight of beds and frames, and a disposition to displace the narrow, many-stepped cone pulleys of the ancient yesterday with others having faces somewhat broader. But there was much hesitation in the bringing out of machines newly de- signed, with the problems presented by the new tools as the basis of THE NEW MACHINE REQUIREMENTS 277 many departures from tradition in the working out of general form as well as of details. At the present time, however, it is possible to obtain machine tools of almost any type which will measure up to the maximum requirements of the new tools, and which possibly surpass them in some cases. This being so, it is pertinent to inquire as to the elements which should be considered in the selection of machines under the new regime. Fig. 239. A drilling machine that fulfills the most exacting requirements for a high-speed, high-power tool. Made by Baker Bros., Toledo, Ohio. Producing Repetitive Work. — Before attempting to indicate some of the most important of these, it is pertinent to point out that the ques- tion of machine tools may well be looked at from two very different view points by the two classes of machine users to whom it might be supposed to be of interest. The requirements of the shop doing little but repeti- tive work, the reduplication of pieces in great numbers, pieces which need to be " good enough " merely (or even which need to be of extreme 278 HIGH-SPEED STEEL accuracy, though possibly to a somewhat less degree), will be much simpler and perhaps less exacting than those of what might be called the general job shop, where a given machine may be called upon to per- form a great diversity of operations upon a large variety of different kinds of pieces. In the former the use of the jig and similar devices to hold the piece operated upon and to guide the tool to insure accuracy, makes possible in most cases the satisfactory use of machines fulfilling two essential requirements: ample powering and sufficient rigidity or strength. Even the latter is essential not so much for reasons of accuracy as to insure longer life to the tool through the elimination of vibration. Fig. 240. An interesting effort to meet modern requirements in repetitive production. The Foster ring-turret lathe. The small range of work generally done upon a given machine, under the conditions named, not only eliminates the need for a great variety of speed changes (once the proper speed for the jobs to be done upon it has been determined), but actually makes such a variety a needless source of expense and something of a nuisance besides. The tendency, in shops having a very large output at any rate, is toward the highly spe- cialized machine, designed specifically for the performance of a single operation or a very few specific operations, on particular parts. For the single operation machine but a single speed and but few adjustments are required; while for the other, the speed changes may well be closely restricted. THE NEW MACHINE REQUIREMENTS 279 The Case of Engineering Works. — Manifestly the shop turning out a limited number of pieces of a kind and doing a large variety of work, whether it be the tool room of a big factory, the general manufacturing shop, or the small jobbing shop, requires a different class of machines. Since jigs and rigs are of necessity but little used, precision must be obtained through accuracy in the operation of the machine. For this reason the machines must meet the additional requirements of extreme rigidity, considerable range of adjustability, and likewise large range of speed variation. Whatever the type of machine, the considerations here pointed out apply with more or less force to all of them. Solidity and Rigidity — How Secured. — Much has been said, and not a little written, about the so-called " anvil " as opposed to the " fiddle " principle in machine design; and it was pretty well established even before the advent of the new tools that solidity, or rather rigidity, is a prime essential in machines for metal working. Solidity and rigidity The Engineering Magaz Fig. 241. How the disposition of material in structural forms affects strength and effectiveness in resist- ing strains. A and B are three and four sided prisms. C is in the form of two girders connected at intervals by girts. B has 10 times the torsional resistance of A, and from 6 to 13 times that of C, the ratio depending upon the strength and frequency of the girts in C. are by no means the same thing. Mass of course does involve inertia, and likewise rigidity. On the other hand it is possible to secure com- parative freedom from vibration without the heavy massing of material often found, if the material be properly distributed. It is well known, for instance, that the hollow cylindrical form of construction is very much stronger in every way than the solid, utilizing the same amount of material; and this fact is made use of in a great variety of ways — - nearly every way, one might have said until very recently, except the construction of machine tools. The hollow prism form also has great advantage over the solid, considering the amount of material involved. It has been shown experimentally that a hollow four-sided box of this rectangular prism shape (Fig. 241) is more than six times as rigid with respect to twisting strains as the same amount of material in side plates with cross girts of the customary proportions. Even the very best possible distribution of material in this form (beams and girts) does not reduce the disproportion more than half. Consequently the very best lathe bed designed on conventional lines can have a strength 280 HIGH-SPEED STEEL ranging only from possibly a fourth (making a very liberal allowance) to an eighth of the rigidity it might have if the material were distributed in the box form. Now the chief business of not only lathe beds, but of the frames of nearly all machines designed for rotating the work or the cutting tool, is to resist such twisting strains; and principles of rational design should indicate the advantage of this simple mode of reducing the weight of metal required, or rather of securing the maximum of strength and rigidity from the metal used. The nearer, therefore, a lathe bed Fig. 242. A heavily girted lathe bed. approaches this form, the greater its efficiency. The same thing holds true of such parts as the cross rails of planers, multiple mills, and the like machines. These most frequently have been made trough shaped, with a section resembling a box of three sides rather than four — a form lacking in resistance to torsional strains, and only about a tenth as strong otherwise as if one-third of the metal were distributed in the form of a fourth side. Weight Essential. — It should not be inferred that lightness is in any wise desirable in machines designed for using high-speed tools. The heavier the frame, that is to say the greater the solidity, the better the absorption of vibration such as inevitably occurs in heavy or rapid cutting; and the better this is absorbed, the greater the efficiency at the cutting edge. The point is that there shall be ample weight, and that this weight shall be distributed for greatest effectiveness in resisting the kind of strains to which the particular machine is to be subjected. And this applies no more to lathe than to milling machines, planers, and the rest of the tribe of machine tools. THE NEW MACHINE REQUIREMENTS 281 Lathe Heads and Others. — Next in importance to the bed, frame, or body of a machine, is the driving or cutting head. Obviously it ought to be in strict proportion to the rest of the machine. It needs strength and Fig. 243. "Lo-Swing" lathe (rear view) of the Fitchburg Machine Works. Unique design of bed — and indeed of machine throughout. rigidity just as much as the body does — but no more. The practice of putting abnormally large heads upon machines otherwise of moderate resisting power, as was done to a considerable extent at first, by no means makes a high-speed machine of an ordinary one. There is no more reason for putting a tremendous head upon an ordinary machine, producing a megacephalous monstrosity ("hydrocephalous," somebody has facetiously remarked, possibly in reference to the designers rather than to such machines themselves), than there would be in using an immense tool without a suitable" support. The head must have solidity, however, and must be large enough to allow for the increased size of bearings necessary for the higher speeds and heavier driving; and in case gears take the place of pulleys, to accommodate the change gears requisite to the type of machine. It is important that the attachment of the head to the bed or body be such as to insure the greatest possible stability; and this precludes, except in machines of special type intended for special service, any adjustability of these parts. A satisfactory attachment is of course possible if the contiguous bases are sufficiently large, well fitted, and securely bolted. No such attachment, however, can be as rigid as if the parts were made in one solid piece; and some of the best 282 HIGH-SPEED STEEL types of machines now have frame or bed and head cast in a single piece. This naturally involves larger and more complicated castings, and it Fig. 245. Section through Darling & Sellers lathe bed and carriage. would be interesting to learn if the same results might not be obtained by casting in two pieces (or more) as heretofore and making them homo- THE NEW MACHINE REQUIREMENTS 283. geneous by one of the several processes of welding now successfully employed in other ways. There seems to be no reason why this should not be the solution to the rather vexing problem of how to secure the largest amount of rigidity with moderate cost of manufacture. Fig. 246. Warner & Swasey hexagon turret lathe. Head and bed cast in one piece; direct connected motor drive; flat turret carriage. Proportions of Bearings. — Properly proportioned bearings are much more important in the new machines than formerly even. Not only is the highly increased speed to be considered, but also the very largely increased pressures. To meet these conditions the bearing surfaces must be large, and of material best calculated to resist wear and to avoid friction. Considerations of space as well as of rigidity make it preferable that the augmentation of bearing surface should result from increased diameter rather than greater length. This permits also the use of hollow spindles, which again tend to greater strength and rigidity and permit the better use of gears in the driving mechanism. Lubricating Devices. — In machines of the types heretofore standard the matter of lubricating bearings (and gears, when used) has not been particularly troublesome. Under the new conditions it is less simple. The tremendously increased pressures and friction make it necessary in many, if not in most cases, to provide some means to insure free and certain lubrication. This involves often the designing of special devices. An ingenious example is that used on one high-speed lathe, Fig. 250, wherein the bearing is surrounded by an oil-filled well. Oil is con- tinuously dipped and carried in sufficient quantity to the highest part of the spindle and is conveyed to all parts of the bearing through the 284 HIGH-SPEED STEEL customary oil grooves. A glass tube inserted in front and properly- protected serves as a level indicator. In another instance (Fig. 247) the gears are about half immersed in oil contained in the pan formed by the gear case. Of course both are well covered. In the case of certain Fig. 247. Headstock of hartness fiat turret lathe. Cover removed to show gears and oil pan. The gears run in oil. very heavy duty machines it has been found desirable to force oil into the bearings by small force pumps. Secondary or Tail Stocks. — In machines requiring a secondary stock for the support of the work turning; on centers, as in the case of the If^CHINCkY NY Fig. 248. Making a lathe center with a high-speed steel cone. The welded spindle with bur, and same completed. Courtesy Thomson Electric Welding Company. engine lathe, this tail stock must be designed with reference to heavy service; that is, with sufficient base, heavy weight, and security in fastening to be in balance with the head stock. It is desirable that THE NEW MACHINE REQUIREMENTS 285 there be some provision for positively bracing it against the bed. The tail spindle, and indeed all centers subjected to much wear, should by all means be of high-speed steel. It - is not necessarily solid, however, for it is sufficient if a high-speed steel center end be welded or otherwise securely attached to a tool steel shank. In large machines it is desira- ble that the tail stock or secondary head be moved as required by an auxiliary motor. Tool-Holding Requirements. — Much attention has heretofore been given to securing a great variety of adjustments in the mechanisms employed to hold the tool and to bring it to its work. The result is that compound rests have not infrequently been " fearfully and wonder- fully made," in order to give a maximum of adjustment and movement. Fig. 249. Method of attaching back rest and carriage in "Lo-Swing" lathe. While of course refinement of adjustment and facility in bringing tool and work together are desirable in high-speed cutting also, it is still more desirable that they be brought together as rigidly as possible. Every joint means a reduction in rigidity; and in work requiring freedom from this to the extent necessary when working high-speed tools at a high efficiency, evidently the fewer slides and other adjustments, the better calculated is the machine to do its work with the minimum of vibration. Absolute freedom from vibration is not possible in a cutting machine. The absence of chattering or of tremors capable of being sensed does not necessarily indicate that they are not present to some extent. At the best they can but be minimized. The tool-holding device and its adjuncts must compare, in rigidity and strength, with the rest of the machine, and require wide bases combined with the least possible altitude. It is not at all essential that the shank of a tool shall 286 HIGH-SPEED STEEL be horizontal, in the case of a lathe, for example; so that there need be no difficulty in considerably lowering the position of the tool with refer- ence to the working center. The same consideration holds in the case of all other machines. Thus that milling machine is best adapted to high-speed work (other things equal) which has short arms and holds the tools closest to the main frame; and in planing, the least torsional strains are thrown upon the cross rail when it is possible to cut with the tool point very nearly in front of the rail rather than below it. It may be added in this connection that the sharp angle or unusual rake per- mitted in the case of one well-known special lathe is in large measure due to this practical elimination of torsion in the bed, the reduction of slide movements to the irreducible minimum, and the lowering of the tool to the greatest possible extent. Reciprocally the tool reduces the pressures and consequent strains, and there is insured better work with lower power consumption than often is the case. Increased Power Required, not Waste. — The powering of the new machines, providing them with pulling power adequate to the needs of the new situation, has developed to a satisfactory state, though not infrequently mere modifications of former types have been attempted in the effort to remodel old designs — generally with small success. It is understood, of course, that increasing the amount of cut, in metal cutting, increases the power consumed; and also that increasing the cutting speed does the same. The increment, however, is by no means proportional to the greater amount of metal removed, and it by no means follows that because a machine consumes power rapidly, energy is going to waste. It is demonstrable beyond question that an efficient, high-powered and high-speed machine using a tool of the new type has an efficiency very much greater than the old machines even if the absorption of power alone be considered in relation to the amount of work done. In other words, on the basis of time consumed and metal removed, a strictly modern machine uses considerably less energy per unit than do the former standard machines. The Belt-Driven Machine. — Of course extremely heavy cutting is necessarily confined, in large measure, to shops doing a particular class of work (comparable to armor plate and gun making operations, say), and have a relatively small place in the ordinary run of general manu- facturing. Nevertheless, the effective use of high-speed steel in this very kind of general manufacturing requires high-powered machines also. Obviously heavy driving power is out of the question in con- nection with the multiple cone pulley, even if nothing be taken into account other than considerations of space and the danger connected with the shifting of rapidly running belts. If a belt drive is necessary, that machine should be best suited to the new requirements which has but few steps in the driving cone pulley (say not to exceed three), or THE NEW MACHINE REQUIREMENTS 287 which has no step cones at all. In the latter case all variations in speed are of course obtained through a variable speed countershaft, a gear box, or the two in combination, the number of changes required being determined by the nature and variety of the work for which the machine is desired. In general manufacturing, as already intimated, a great variety of speeds and feeds is not only unnecessary but often a source Fig. 250. Lodge & Shipley single-face pulley drive in connection with gear box. Cover and bearing caps removed to show positive lubricating device. of needless expense and annoyance. Wherever possible, it is better to have the machine drive suited to a small range closely accorded with the requirements of its special work. Few Speed Changes Necessary. — The gear box in connection with a variable speed motor offers a combination with which are obtainable a maximum number of speeds in cases where the work of a machine is much diversified. Even in such instances the number of different speeds really necessary need not be very great. The new tools lend themselves admirably to wide latitudes in respect of speeds, and feeds also; so that in spite of an evident tendency in some directions to require more rather than fewer changes, it would seem the better plan in general to get along with the less number, except under special circumstances. Amount of Power Absorbed. — The amount of power consumed by machines under former conditions has been very generally overestimated because, among other reasons, the proportion absorbed by line and 288 HIGH-SPEED STEEL countershafting and other transmission devices has been underesti- mated. Except in case of very powerful machines the energy absorbed per unit, while cutting, rarely exceeded one horse-power. The newer machine tools of similar capacity require rarely less than three or four horse-power, and not infrequently several times as much. A 20-inch swing lathe, to take a concrete example, running at high-speed and taking a heavy cut, has on occasion absorbed a maximum of 30 horse- power, though its average consumption is less than 10 horse-power, and the minimum a good deal less than that. In turning operations, under favorable conditions, the power absorbed per pound-hour of metal removed varies from 0.03 to 0.07 horse-power, over and above that required to drive the machine itself. This is measurably less than under former conditions, where the consumption has been close to 0.05 or 0.06 horse-power and upwards. Belts for Auxiliary Drives. — Using up power so rapidly, pulling loads as heavy as are thus required at the high speeds conditioning the work, 1 ■ V ■ v T% [ ■PI ^PJ 1 "V_ A gggfw • "^B Fig. 251. A heavy milling machine with auxiliary motors for movements other than the main drive. means practically the elimination of belt drives for the auxiliary move- ments of a machine, and the modification of the main drive also, as above pointed out, with the tendency apparently toward a main drive pulley, where this is used at all, of but one belt face, and that driven THE NEW MACHINE REQUIREMENTS 289 whenever possible by a silent chain rather than by a belt. The speed changes and the auxiliary drives, therefore, are through gears; not only in the main drive, but in the feeding and other movements wherever these are necessary. The gears are put under such stresses that the ordinary ones are quite unsafe, and only steel or bronze is permissible as a material in view of the limitations necessarily imposed as to size. Steel rimmed gears also are satisfactory in large sizes. Of course steel gears are positive — too positive for the safety of the machine, it is sometimes urged; but shearing pins or other safety relief devices obviate any possible objection on this score. It is important to bear in mind that the feed- ing stress may under abnormal conditions (as of a very dull tool, for example) quite equal the driving stress; and it is therefore essential that the feed mechanism be, if not just as powerful as the drive, certainly adequate to the probable exigencies to be met in the operation of the machine, and therefore much stronger than usually found under former conditions. It is desirable also that all high-speed machines be pro- vided with good braking devices for quickly stopping should occasion require. Relation of Powering to Capacity. — It seems to have been a pretty generally recognized principle of machine design, prior to the advent of the high-speed era, that the massiveness and powering should increase proportionately to the capacity, or more precisely, the size of the piece the machine could accommodate. If there ever was any reason for this, there is none in the case of the new tools. Quite evidently a machine, say a lathe, required for working down a 4-inch bar, may be required to cut just as fast and to take as heavy a cut as if the bar were 12 inches or any other large diameter. The same gearing and powering obviously, then, are required for each. The exception of course is where the smaller machine is required for work which actually is lighter in character and which consumes less power. Even in such cases it is well to be on the safe side and to insist upon weight and power commensurate with that commonly found in the machine of larger capacity. The point is that the machine be capable of taking off and using continuously the largest amount of power that can be efficiently utilized by the tool. The Electric Drive. — While the individual motor drive has many manifest advantages in ordinary practice, its efficiency is not neces- sarily superior in all sorts of machine operation. In high-speed cutting, however, the largely increased power consumption and the desirability, in general manufacturing, of quickly stopping and starting machines when changing the piece under operation, strongly emphasize the dis- advantages of line shafting. The power lost in line transmission under ordinary conditions is very generally underestimated. Under the new conditions, wherein usually increased weight and speed is required in both shafting and belts, the losses are still greater and the individual 290 HIGH-SPEED STEEL (or at any rate the group) machine drive comes nearer yielding a high efficiency, unless possibly in the case of small machines. Even in the latter case the group drive is preferable to line transmission of power. The method of applying the motor to the machine varies greatly, ranging from a separate motor at the ceiling, floor, or other convenient location, driving directly by a main belt or indirectly through a countershaft; to mounting it directly upon or building it around the main driving shaft of the machine. The latter is probably the most effective method of hitching, especially when the motor is of the multiple speed type; though the advantages of a chain drive through a variable speed countershaft Fig. 252. A Lodge & Shipley high-speed lathe, direct motor connected. or other device are to be considered where it is desirable to provide a great range of speed variations. The attachment of motors to machines through bracket extensions should be avoided. Use of Auxiliary Motors. — A development of the motor drive which is of striking advantage, especially in big machines, in that it does away with a number of complications arising from effecting auxiliary move- ments through the main drive, is the employment of subsidiary motors for the latter purpose. Thus we have a lathe, Fig. 238, wherein the move- ment of the carriage is effected by a small motor, which latter also traverses the movable secondary headstock. Likewise a large planer, Fig. 254, is provided with as many as five separate motors. The main motor actuates the drive, a second elevates the crosstail, a third traverses the head and gives vertical movement to the tool, another operates the side heads, and so on. The simplification thus possible, and the convenience of operation, are ample justification for the innovation. Limitations of Reciprocating Machines. — In wood working and the like operations, rotating tools or machines are about the only ones in use. There is everywhere a distinct tendency away from reciprocating THE NEW MACHINE REQUIREMENTS 291 o.g >> 13 ■- .« 3 a ^ o "a o sj3 o ** 292 HIGH-SPEED STEEL tools, with their great losses of energy through reversing, and toward the larger use of machine tools which rotate either the work or the tools. It doubtless will be a long time before the planer, the shaper and the slotter are entirely displaced by rotary planers and milling machines. The day is, however, being hastened by the apparent impossibility of designing or operating these types of machines to work at an effective speed approaching that of rotary tools. About the highest speed commercially practicable with a planer is about 60 feet per minute; and this is not at- tainable in general practice under normal conditions. If greater efficiency is to be attained it will doubtless be through the perfection of clutches and a modification of the method of changing the direction and rate of motion at the beginning and end of each stroke. Already progress has been made in this direction, and planing machines are now designed so that the tool starts into the work at a speed which does not damage either tool or machine, and is rapidly accelerated after entering the work to the maximum permissible in the operation of machines of this type. Taking Care of the Chips. — When cutting speeds and feeds are moder- ate, taking care of the chips presents no difficulties to speak of. When, Fig. 254. Chip breaker (cover removed) as used on Hartness flat turret lathe. Taken by permission from "Machine Building for Profit," by James Hartness. however, they come off so fast, as has been the case in certain instances, that the operator is obliged to exercise considerable agility and caution to avoid being entangled in and burned by the hot chips, or where it requires the services of two laborers to keep the machine clear enough of chips to permit effective operation, as was the case in a certain experi- ence, the problem cannot be disregarded. Unless attention is given to it in machine (and tool) design, steel chips are liable not only to injure THE NEW MACHINE REQUIREMENTS 293 _d o > c« fa >> 294 HIGH-SPEED STEEL the operator, as in the extreme cases cited, but accumulate with such rapidity as to be otherwise troublesome. Most of the difficulty arises in the case of steel turning, by reason of the length of the curled chip. This feature is very simply cured by some device attached to the tool or tool post which bends the chip as it comes from the tool so it is broken into short lengths, say two to four inches each. An adjustable chute or other conveyor to carry away the short pieces is desirable. Summary. — The remarks here made with reference to features in machine tool design desirable in connection with high speed steel tools, are applicable (except as specifically indicated to the contrary) to practi- cally all types, whether used for turning, planing, milling, boring, drilling, or what not. The points to be emphasized may be summarized thus: Fig. 256. Work of the chip breaker. The six groups of short chips were produced by a chip-breaking turner. The fine, curling chips were not run through the breaker. Taken by permission, from "Machine Building for Profit," by James Hartness. Ample weight and distribution of material to insure the maximum of rigidity. Elimination of all joints, movements and connections not absolutely essential to the kind of work to be done. Locating tool as near the base (frame or bed) of the machine as possible. Strengthening of main and subsidiary drives and substitution of posi- tive powering for uncertain belts and step cones. Reduction of speed changes to the minimum required in the special class of work demanded. Direct individual motor drive wherever feasible. Provision for taking care of chips. . CHAPTER XIX. REMODELING AN OLD EQUIPMENT. To Scrap or to Remodel. — A well-established canon of industrial engi- neering is that if a new. machine will save its cost in five years, it should by all means be installed and the old scrapped. Not a few concerns make it a regular practice to supplant equipment, even when com- paratively new, with other if a saving half as great can be shown. If this principle were to be closely followed in all establishments, there would now be a great amount of second-hand machinery easily obtain- able. In many cases the saving to be effected by the new tools is insufficient to warrant scrapping existing installation. Under these conditions, and likewise where managerial conservatism or expediency in reference to capital required, make it necessary to use old machines with the new tools, it is important that some attention be given to the matter of remodeling them and bringing them into better condition to get good work out of the tools. Many a machine, in a general manufacturing shop at any rate, can by remodeling, or even by some minor changes, be brought into a high state of efficiency under the new conditions. A Case in Point. — For example, a large capacity lathe, or rather one of large swing, is required to take light cuts only. Its weight, strength, and pulling capacity ordinarily will be quite sufficient for a very con- siderable increase in speed, with possibly a modification of the driving pulley; and by similarly changing the cones of the feed drive, or better by substituting gears, a considerable increase in feed traverse also is obtainable. Such a modification practically makes a high-speed ma- chine of an ordinary one, adapted to the particular work mentioned, and is possible in a good many cases. When undertaken, however, it is important that the bearings be suitable to the new conditions. Other- wise it may be necessary to put an entire new driving head upon the machine. Simplification in Remodeling — Powering. — Such a new head, or possibly a modification of housings and accessory parts in the case of machines of other types, must be designed not only with heavier bear- ings, but with provisions for lubrication both ample and certain. Unless this is done vibration and excessive wear, with the resultant chatter- ing, are a natural consequence. Under the circumstances mentioned most machines now in use are stiff and solid enough to stand a mod- 295 296 HIGH-SPEED STEEL erate amount of heavier duty, if the driving parts are made propor- tionately as effective. This is particularly true in the case of general manufacturing, where the work is mostly repetitive, the pieces of mod- erate length only, and the tools or work guided by jigs and other like devices. In such cases the re-designed driving parts can be very much Figs. 257 and 258. A drill press remodeled, to adapt it to high-speed drilling. Wheeler Company. Courtesy of Crocker- simplified by the elimination of all superfluous speed gears and move- ments intended to give facility and convenience in miscellaneous work. The expense of thus remodeling a machine necessarily used in general jobbing is not infrequently prohibitive, and about the only thing that can then be done is to give a little increased speed and feed, to the limit of the machine's capacity; and perhaps to increase the driving power some- REMODELLING AN OLD EQUIPMENT 297 what by the substitution of broader and larger driving cones. In the case of motor driven machines the motor should be, if it is not, capa- ble of carrying overload sufficient to meet the requirements of the inter- mittent increase in its work. Fig. 259. Increase in size of feed gear on an old lathe. Due to increased feed possible with, the new steels. One of many ways in which old machines may be rebuilt to meet the new conditions. Fig 260 Effect of increasing materially the pull upon a planer not designed for high-speed service, and a method of prevention. Steel gears are preferred, but in this case a steel nm was shrunk upon an iron hub as an emergency measure, and worked all right. Courtesy of Mr. tt. W. Jacobs. Special Attention to Gears.— In overhauling a machine with a view to use with high-speed tools it is essential that all gears should be replaced which are much worn or which are not both truly made and strong enough for the new service. In many cases it may be necessary to replace cast-iron gears with bronze or steel ones, or where they are large 298 HIGH-SPEED STEEL enough to warrant, with steel rimmed (and preferably also steel or bronze bushed) wheels. Especially in reciprocating machines like planers, any large augmentation of the work done is likely to be accom- panied by a stripping of gear teeth during the reverse and the starting of the tool into a fresh, cut. The stopping and the sudden starting absorbs an amount of power frequently not suspected; and the momen- tum is greatly increased by the additional speed — where the design of the machine will permit such increase. To provide against stripping under these circumstances the gears must be strengthened as already indicated. Modification of Reciprocating Machines. — Sometimes a machine of this reciprocating type, not otherwise capable of increased speed but strong enough to stand it, can be better adapted to the new tools by a re-designing of the driving apparatus and the use of clutches or the modification of those already in use. At the least it is usually possible to increase cuts, even if not speeds, through such modifications. As to Worn Machines. — Concerning the many machines likely to be found in a shop devoted to general manufacturing, machines likely to have been in service for some time and therefore more or less worn, not much can be said without an understanding of the particular conditions. Often nothing is possible except to take advantage of the longer wear of a tool, that is, the less frequent need for grinding, while running at the usual speeds, though possibly with some increase in feed. In other cases it is possible, and therefore expedient, to run at the highest speeds possible without remodeling the machine, getting all possible out of it within the shortest time, then scrapping it when too much worn to work with sufficient precision and freedom from vibration. Spring in Frames. — In machines like drills, and even planers, espe- cially when the housings are considerably separated to allow large pieces to be worked, there is likely to be more or less spring in the rail or in the parts bringing the tool and the work together. This is bad enough when ordinary tools are used, and is the frequent cause of tool breakage, especially of drills; but if advantage is to be taken of the higher speeds and feejis often possible, attention must be given to so strengthening these parts or so supporting them that there shall be no spring. If the frame of such a drilling (or other) machine is weak, little, if anything, can be done. If the trouble lies in worn bearings for the brackets supporting the table, or to spring in them, they can usually be solidly blocked against the base and the proper degree of rigidity secured. Spindles and Tool Position. — It should go without saying that worn spindles are not permissible in high-speed work; and if there is no pro- vision for taking up the wear, they would better be replaced. The same thing is true of all studs and other bearings, and applies as well to all slides and adjustments. In many cases these need strengthening as REMODELING AN OLD EQUIPMENT 299 well as refitting, and usually also may be modified so as to bring the tool position nearer the supporting frame or bed and thus decrease the liability to chatter. All slides and adjustments not absolutely neces- sary should be eliminated and the tool holding and adjusting parts made as rigid as possible. CHAPTER XX. STATEMENT OF THE PROBLEM. Position of High-Speed Tools. — Revolutions do not usually happen in a moment, in the industrial world. Perhaps it were better to say, they do not happen at all; for industrial progress is evolutionary rather than revolutionary. A new invention does not immediately upset prevailing conditions, but gradually takes its proper place as an economic factor while it is being developed and adapted to existing conditions — or as not infrequently happens, while conditions are changed to meet the new order made possible. So it has been in the case of high-speed steels. Their capabilities and limitations are by no means definitely fixed even yet, though they have already taken a place in production engineering important to a degree scarcely equalled by any other discov- ery in recent years. Along with improved methods of rapid transport and handling of materials, and of rational organization and administra- tion, the new tools are surely, if not precipitately, bringing about a new order of things in the metal industries, and to a considerable extent also, indirectly, in other industries apparently not closely related. Cost Reduction as High as One-Third. — A manufacturer of large interests is quoted as saying that in his plant the cost of producing machines has been reduced a fourth to a third, directly through the economies permitted by the use of high-speed steel tools and the indirect economies necessitated by the reorganization of the shop methods and shop administration. This may be an unusual case; perhaps it could not be duplicated in a manufacturing plant producing a greater variety of output. Perhaps also it is not possible in many cases (though parallel instances could be mentioned) to find, as in this one, that the labor-time on a single piece could be reduced as much as 95 per cent, the product being at the same time better than before. The Case of Tire Turning. — The turning of locomotive drive wheel tires, to mention a case under different conditions, likewise is indicative of the wonderful development in metal cutting possibilities. Not so long ago one and a half to two days was the time commonly consumed in turning up a pair of drivers; and very recently indeed, the perform- ance of the job in half a day, or a little less, was looked upon as a feat worthy of attention. Now, however, it is not only possible, but it is a regular commercial performance, to turn up badly worn tires as large 300 STATEMENT OF THE PROBLEM 301 as five feet in diameter, at the rate of a dozen or so in a day of ten hours — the time required for placing and removing the wheels included. Fig. 261. Turning locomotive drivers at the rate of a dozen or more pairs a day necessitates facilities for rapid handling, as here shown. Astounding Economies not to be Expected. — The publicity matter of high-speed steel makers and sellers sets forth in glowing terms examples of wonderful performances and astonishing savings effected. There is no need to question the accuracy of the instances cited. Doubtless even the most extravagant is capable of verification; and certainly none is likely to be more wonderful than those mentioned above, or than such as might be found in almost any average shop. Manifestly, how- ever, such remarkable savings are not possible in all cases, nor perhaps even in many, considering the great diversity of conditions affecting 302 HIGH-SPEED STEEL the different kinds of jobs in general manufacturing, and the differences in efficiences everywhere existing. Indeed, the authority just quoted particularly mentions that in his own business of machine building, roughing work, in which he makes the greatest showings, does not exceed one-third of the work required, and that consequently a very large part of the saving is on this third part. It is not at all necessary that such showings should be common. The ordinary performances of the new tools, even under conditions allowing only moderate efficiency, are marvelous enough. A Modern Miracle. — The possibility of cutting refractory metals almost as if they were cheese is another of the modern miracles — a miracle already become commonplace. Only recently cutting at something like 20 feet a minute was considered reasonably satisfactory; and this rate was probably above the> average in most shops, though indeed speeds as high as 60 feet per minute were not unheard of, and some considerably higher than that are recorded as having been attained. So far as can be learned the maximum speed attained with carbon steel tools was about 100 feet per minute; and this was under the most favorable con- ditions of cooling and chip removal, conditions involving a complicated system of appliances and devices scarcely applicable to ordinary com- mercial work. Half as great a speed may well be considered to have been the commercial limit, under the old conditions — a limit not attained in practice to any considerable extent. The Passing of Traditional Conditions. — Such speeds, or rather such extreme lack of speed, was in a sense distressing in an age where rapidity, time-saving, nerve-racking haste has come to be characteristic — where seconds count as hours did not in times within the memory of most of us. The days are past, or at any rate are swiftly passing, when it is the regular thing to see a piece of metal lazily creeping round and round, the tool paring it down at a snail's pace; the operator the while listlessly lounging near, merely keeping an eye upon the machine to see that all is going properly. And it is well that it should be so. Efficiency is come to be the watchword of modern civilization, and of industry the slogan. Hence the new tools, or possibly newer ones of still higher possibilities, must eventually crowd out the less efficient wherever efficiency counts, and must at the same time bring about very great concomitant changes in all the conditions touching or involved in the metal working arts. But these changes are not yet accomplished, taking productive in- dustry as a whole. They are only in process, and completed, relatively speaking, only in isolated industrial units. Hence it is worth while to inquire into the situation, not so much perhaps to discover why high speed steel tools are as yet used so little, comparatively, as to determine if possible what problems are to be solved, and how, in order to take the STATEMENT OF THE PROBLEM 303 fullest advantage of the new steels consistent with expediency. So much, at least, should be done in all cases. Only that ultra conservatism which spells bad management would permit less. Non-Uniformity of Material. — A serious difficulty in not a few shops is that connected with the lack of uniformity in the hardness of the material to be machined — a difficulty which sooner or later is en- countered in every shop. Castings every once in a while, for some unaccountable reason possibly, come so hard as to play havoc with ordinary tools. Occasionally the same thing happens with steel stock. Only a few pieces at most can be finished until the tool requires re- grinding. Possibly, in extreme cases, several grindings are necessary in order to finish a single piece. Or it may be the stock specified and required in a given part, is uniformly so hard as to be practically beyond ordinary tools — even mushet steel being able to make but a sorry show- ing. Instances of this sort are by no means infrequent; and there can be no question as to the expediency of using high-speed tools, even though there be no gain in speed or cut — as most often there may be, never- theless. The gain here will certainly be great if only the loss of time in grinding and setting tools be taken into account; and it is likely to be considerably augmented by the saving which arises from the elimi- nation of the need for scrapping many parts because of inaccuracy such as necessarily accompanies those conditions, when only carbon steel tools are available. Especial Field for High-Speed Tools. — The especial field for high- speed steels, or rather the kind of work in which it shows up most favor- ably, is that involving the removal of large quantities of metal, as has been pointed out elsewhere. Here the Taylor doctrine of running tools (these having been suitably designed, standardized, and treated,) at speeds high enough and feeds and cuts heavy enough to necessitate re-grinding at intervals of about one and a half hours, can be practiced to the best advantage; and here it is that there is most reason for his dictum that the one who does not do so, does not know how to cut metals most efficiently. Scrapping not Generally Warranted. — Saying nothing of the great variety of work where heavy cuts are not only unnecessary, but undesir- able, and where the highest speeds are impracticable, it is to be observed that the conditions of maximum effect are dependent not alone upon the tools, but upon a considerable number of concomitants, the most important of which perhaps is the machine equipment. Obviously it is desirable that this latter should comport, in possibilities, with the tools as applied to the particular jobs. But average machines as hereto- fore used, and still in use for the most part, by no means measure up to this standard. In a few instances concerns have adopted the radical policy of replacing entirely all equipment not capable of using the new 304 HIGH-SPEED STEEL tools to their maximum capacity, with other that is so capable. Clearly such a policy is in line with " good business " when the economies to be effected are great enough to warrant the expenditures. But in miscellaneous manufacturing it seems quite certain that they are not always so. Take for example that very large class of operations upon small pieces reduplicated in large numbers and generally requiring but little machining: The limit of the operator's endurance, and therefore the limit of output (except possibly through the adoption of an auto- matic machine — which is as yet impracticable in the vast majority of cases), has under these circumstances usually been reached. Place of the Automatic Machine. — Of course if an automatic machine, or even a semi-automatic requiring a minimum of attention and skill from the operator or attendant, is feasible — which is to say, if such a machine can be built without being very complicated and expensive to maintain — the problem changes again, and it would be desirable to build the machine. But whether this be feasible or not in particular cases, there is a side to the high-speed problem often overlooked, in this very matter of the endurance limit of the operator and the related psychological and sociological effect of the deadening monotony involved in feeding stock into, and practically becoming an attachment of a machine. This is not deemed the place for a discussion of this aspect of the new-tool problem; but it is an aspect which will year by year become more insistent for solution and which must sooner or later be squarely faced. And when that situation arises, the indications now are that the increased development and use of the automatic machine, with its large possibilities in the way of high-speed tools, will be an important factor in the ultimate solution. Limitations Imposed. — Not only is equipment wanting, in the great majority of cases, but expediency prevents the scrapping of machinery still in good, or even in moderately good condition, and the consequent large expenditures for new. On the other hand also there are a good many jobs where the inherent conditions are such that the machines in use are quite competent to do all, or nearly all, that would be possible in any machine — where the efficiency of a cutting tool is to a consider- able extent limited by the nature of the job itself. Take, for example, the machining of a heavy casting, where the machining itself amounts to little and occupies but a small fraction of the total time necessary for the complete operation. Increased cutting speed manifestly would be of no considerable advantage; and neither would heavier feed; like- wise there could be nothing gained by deeper cats, except in special cases, for these would merely involve molding the casting larger than necessary, for the mere sake of removing it again. The special cases would be typified by that wherein it is required to mold the casting considerably heavier than the finished size in order to minimize distor- STATEMENT OF THE PROBLEM 305 o a u IS a bo 2.9 p to '5b S O M 2.S o 2. 29 a m C3 c3 •s O O 0) 306 HIGH-SPEED STEEL tion or possible breakage during a series of operations, or where the nature of the casting partly chills the surface to be machined or leaves it with a skin naturally hard on a cutting tool. Here a deeper cut would be desirable; but the machines already installed generally would be quite able to take care of the increase required. Two Extreme Classes of Jobs. — From jobs of these sorts it is a far cry to those at the other extreme where long and heavy cuts and high speeds are obviously the thing; and between these extremes lie jobs of all gradations as to cutting possibilities. Lying in the lower ranges would be those where the only economy in high-speed tools would be the lowered cost of tool and tool maintenance, and those where under prevailing or former conditions many pieces were necessarily scrapped because of breakage or imperfect workmanship. The apparent paradox of cheaper tools is touched on elsewhere, but may be said here to refer to the relatively long life of such tools, the more particularly of com- posite tools, and the consequent distribution of tool cost over a very greatly increased product. The scrapping of many pieces which has seemed necessary to and inherent in tools and methods heretofore in use, can be eliminated to a material extent through the use of high- speed tools, in most instances without any change in machines. Such a saving may easily be considerable while still not appearing in a changed labor rate. Operations with Short Cutting Times. — Referring again to jobs such as have just been mentioned, where the cutting time is relatively small: there may be many such where it is quite possible to use higher speeds or heavier feeds to such an extent that the cutting time becomes negli- gible, or nearly so ; and thus the second or third machine which is under present, or was in recent practice necessary to keep the operator busy, can be dispensed with and capital (and consequently depreciation, repairs, etc.) thereby reduced while at the same time available floor space is increased. Here also there would be a saving quite appreciable, which nevertheless would, under usual methods of cost accounting, not appear in labor cost — a saving which might, or might not, necessitate the installation of a heavier machine, according to the conditions in par- ticular cases. Immediate and Ultimate Considerations. — Evidently the problem is by no means, or at any rate not at all necessarily, one of disposing of old and installing new equipment throughout an average plant. Unquestion- ably high-speed steels, if the standing offer of certain makers to replace with positive economy any ordinary tool with one of their own make can be backed up, are bound ultimately to displace almost, if not quite entirely, the less efficient kinds. Even now it is not so often a question as to whether or not a high-speed tool shall be used, but rather as to expediency in reference to the equipment in which the tools must be STATEMENT OF THE PROBLEM 307 operated. Ultimately of course the question of profitableness will determine, as it does practically all questions in industrial engineering; and if, when all is said and done, a new tool or a new machine, or both in conjunction, will yield a product cheaper and better than before, then the old must inevitably give way to the new sooner or later — and usually sooner than later. For the present, however, the problem, in so far as it concerns the shop or factory in general rather than those special cases already indicated, seems to be less a general than a specific one. In an offhand way it may be said with assurance, almost as a matter of course, that high-speed tools should be used to the largest possible extent in all metal-cutting shops. But the real question is, just what shall be done in this or that specific, particular instance? How the Problem Works Out. — Here is a job, let us say, to be performed under definite conditions; and such and such machines are available for, ■ or possibly are actually performing, the operations. Is it possible to reduce the cost of these operations, or specifically this one operation, by the use of a high-speed tool and heavier or faster cutting? If so, to what extent are the available machines capable of realizing the ideal conditions, if utilized without modification? If this performance falls short of the attainable maximum, can the machine be altered or rebuilt, without prohibitive cost, so as to yield this maximum while still not working disaster upon the machine and wiping out the gain through heavily increased maintenance cost? This latter point is one seriously to be considered, in connection with the subject of equipment in gen- eral, as well as in connection with specific cases. It is found possible to speed up a machine still quite serviceable under ordinary conditions, so as to yield a considerably larger output; but directly it may be found also that gears break frequently, and belting gives out rapidly, and the expense of repairs in general is possibly as much as doubled. This of course does not always take place; but it may do so, and is frequently to be expected. The limit of the machine's endurance, that is, the point beyond which maintenance cost becomes excessive and expensive delays through breakdowns are invited, is carefully to be considered; and along with it the rapid wear under the severe conditions for the meeting of which it was not designed. The Power Problem to Receive Attention. — Furthermore, the matter of power consumption needs attention. Not that the increased amount of energy required for taking care of the greater amount of work done need occasion serious concern. Under proper conditions the total power required for doing a given amount of metal working will actually be less than under old conditions, though indeed it becomes necessary to concentrate or localize it largely, so to speak. In attempting to speed up old machines designed only for moderate speeds, the amount of power absorbed by the machine itself, not considering at all that 808 HIGH-SPEED STEEL entering into the cutting, may easily become surprising. It is not so unusual as might be supposed in the absence of actual measurements, for a half of the total power delivered to a machine to be thus absorbed in overcoming friction. Under such circumstances the speeding up of a whole plant would evidently necessitate a very large augmentation of power plant capacity. When a Machine is to be Superseded. — When it is evident that a machine already installed cannot economically or effectively meet the desired requirements, the question arises as to the displacement of the machine with one capable of yielding the maximum output with min- imum maintenance and operating cost. Questions of temporary expedi- ency aside, there are pretty definite conditions, though varied according to the nature of the individual cases, under which a new machine should replace an old. Quite plainly if the required output involving a par- ticular job or a closely related class of jobs is sufficient to keep the machine busy practically all the time, and the required increase in out- put must be met by additional equipment anyway, it is then not only desirable to install an up-to-date machine, which by its increased effi- ciency will be able to take care of the required increment, and probably more; but it is folly not to do so. Even though the machine be idle a considerable portion of the time under the new conditions, the econ- omy is like to be more than great enough to warrant the change; and the same often will be found true when even the old machine is not used nearly to its capacity. The common rule that a machine is to be replaced whenever a yearly saving or ten to twenty per cent of its cost can be shown, has been referred to already. In the consideration of such changes it is to be remembered that a machine not sufficiently produc- tive or powerful to allow efficient use upon one job or class of jobs, may still be suited to economical work upon another whose requirements are less rigid; so that the displacement of a machine is not necessarily the same as scrapping it. Re-design of Jigs, etc. — Not only must the machine be capable of, and adapted, so far as may be, to heavier duty, generally speaking, but especially in reduplicative work jigs and other holding or guiding devices will need re-designing or remodeling. Magnetic or air chucks and jigs will largely displace the cumbersome lug or screw fastened holders still generally used, so that a piece of work, or a number of pieces simulta- neously, can be instantly fastened securely and held firmly during the operation, and as quickly released when the job is completed. Like- wise it will be necessary to devise something more substantial for hold- ing work turning on centers than the bent tail dog, whose wagging is not tolerable under the new conditions. Manufacture or Purchase of Tools. — Intelligent production and han- dling of the tools is a factor in high-speed production second in importance STATEMENT OF THE PROBLEM 309 to none other. The actual making of tools is best not undertaken at first, except perhaps in large plants. They can be readily purchased made up to specifications fitting them- for the particular work required. Fig. 263. End milling. Use of magnetic chuck for quickly clamping and securely holding work. Courtesy Cincinnati Milling Machine Company. The purchase of all standard tools will be most economical, in general, for all shops except possibly those especially equipped for their manu- facture, though it may well be that such simpler forms as lathe tools and the like can be produced within the plant itself. Even this is not advisable unless there be suitable facilities and tool-makers expert enough to make tools of uniform and high standard quality. Experi- ences with tools manufactured under uncertain conditions and by inex- perienced hands, and therefore of uncertain quality and uniformity, are likely to prove unsatisfactory and disappointing. In the beginning, at least, it is safest to buy all tools ready made, gradually training tool- makers to the proper handling of the new steels and substituting for those made outside only as experience shows the possibility of producing within the shop others equally certain in quality. An alternative of course is the employment of one or more experts to undertake the tool- making problem, and in large shops to train the rest of the tool makers to the new tricks of the new trade, so to speak. Unquestionably this is 310 HIGH-SPEED STEEL an excellent thing to do anyway, in places where many tools are manu- factured; and even then it will be no day's task to educate men brought up under the old conditions to the new requirements in tool making. Expert Direction. — Such an expert might have charge also of the development of the high speed steel problem throughout the plant, with a large responsibility in the matter of educating the machine operatives also, to the new situation. The old proverb holds good in industry as elsewhere — it is hard to teach an old dog new tricks; and it takes time and not a little persistence to educate a man out of the 25 foot speed and sV inch feed rut and induce him to take advantage habitually of cutting rates three and more times those to which he has been accustomed, even if he be willing to learn. The new tools can do much; but they cannot make an industrious workman of a lazy one — any more than they can increase the physical endurance of one already pushed to the limit under the old conditions. Where this limit is reached, as shown before, an entire change in the method of doing the work — perhaps by the substitution of automatic machinery — is the obvious thing to do. If not, it is desirable that the hearty co-operation of the workmen be secured. Attitude of Operatives. — Not that operatives in general do not take kindly to the new tools. On the other hand they all but invariably welcome and eagerly desire to be permitted to use them. But it not only requires an expert knowledge of conditions and of the possibilities of the individual cases so to set the pace or change conditions that the highest attainable efficiency shall be insured; but also it is essential that there be held out, through a rational and just wage system, like the premium plan, say, such incentive as will stimulate ambitious workmen to raise their own efficiency; and that at the same time there be super- vision so skilled and so organized that guesswork in machine operation is minimized or entirely eliminated, and conditions hindering maximum efficiency changed so as not only to permit but to compel it. Maximum Production — Auxiliary Conditions. — Precisely here it is that a great mistake is often made in high speed tool practice. The subject has scarcely yet been approached, much less reduced to definite standards fitting all cases. It is true that Mr. Taylor and his associates have succeeded in reducing practice in certain works to a very definite basis; and they have obtained results nothing short of phenomenal. Equipment, tools, auxiliary methods, and even administration, have been revolutionized to create conditions making for greatest efficiency; and everything (or almost everything, it would seem) is definitely and specifically worked out by the slide rule, and by effective supervision the standards thus set are actually attained and maintained. All this goes to show, as do similar experiences elsewhere (possibly carried out less consistently), that the problem of high-speed tools involves not only STATEMENT OF THE PROBLEM 311 MS O 03 -a m wi3 a o oj-0 to °l MS Si .2* 1.3 o 2 t*-i a a'3 t3 a «> § 2 _-2 «* oJOfe S * **| — a o a.Q-3 ~ a o §2-3 *%£ J3 -^> +3 ■* ^ -2 312 HIGH-SPEED STEEL the matter of tools and machines, but is vitally concerned with the sub- ject of shop organization and necessitates methods of supervision and co-ordination very much in advance of those customarily in vogue. It is important, that is, not only to determine beforehand with practical accuracy the machine to be used for a particular job, and the most efficient speed, feed and cut, and the precise manner and order of doing the work; but for profiting to the largest extent by this possible acceler- ation, the movements of material, the facilities for storage, the supply of sharp tools and the method of distributing them — all these and Fig. 265. Auxiliary storage for materials in process requires much space around the machine or in- volves the large use of conveying or transportation units, as in the case of this overhead trolley system. From the main trolley track switches run to every erecting jack in the room. The assembled machine is tipped from the jack onto the hooks of the trolley, shunted out upon the main track, and propelled to the paint shop by a motor-actuated endless chain provided with suitable fingers. other ancillary activities, systems, and methods concerned in the con- version of material into product, may need, and probably will require, entire reorganization. Thus, concretely, it boots little that an indus- trious and ambitious workman be provided with the most approved superior rapid-cutting tools operated in the most up-to-date machine, the most efficient rates of cutting carefully prescribed, and working under a rational premium or other wage system urging him on to the exertion of his maximum efficiency, if at the same time, he is obliged at intervals to loaf or dawdle along while, waiting for material, because of inadequate means of transporting and handling the same to and from STATEMENT OF THE PROBLEM 313 his place of work, insufficiency of stock or of storage facilities for material, or a stinted supply of sharp tools. Material Handling — Change in Methods. — More than likely acceler- ated production under the new conditions will mean, in most factories, a complete change in the methods and facilities for stock storage, the providing of more room and better access so as to permit handling to and from storage with the greatest convenience and ease. Heavy parts, as well as light, will need to be so stored, and means for mechan- ical handling so provided, wherever necessary, that the movement and Fig. 266. After passing the dipping tank the trolley and its load are switched from the trunk track to storage tracks for drying. handling shall involve a minimum of effort, of time-labor. The subject of auxiliary storage, storage for material in process of manufacture while passing from one operation to another, becomes highly important. More room will be required around a rapid production machine than ever before; but not necessarily for storage bins. These may well be required; but if so, ought to be of such type that they can be emptied into transports of appropriate sort with practically no handling, or to be so placed as to permit their being reached without any transport, by the operator to whom the parts next pass. Auxiliary Storage and Transportation. — The auxiliary storage will most likely need to consist mainly in a very materially increased number 314 HIGH-SPEED STEEL of transport units — cars, trucks, trolleys, or whatever such units may- most conveniently consist of, to meet the requirements of the particular kinds of parts to be transported. Hand trucks, of approved types only, may perhaps still be used, in larger numbers than ever before; but the relative inefficiency of man power, as compared with mechanical, indi- Fig. 267. Auxiliary storage by hand trucks. Very well if trucks are designed to fit the conditions and do not occupy an amount of space not permissible. cates the need for displacing hand trucks to the largest possible extent by conveying units capable of being mechanically moved. The trans- portation system therefore will need overhauling and remodeling, with a probable large increase in capacity and a close interrelation of the various constitutents. The standard gage and industrial railway, the crane service, the overhead trolleys, belt and other mechanical conveyors, and also hand trucks where these are retained, must be so interrelated and efficiently operated that material will move rapidly, without unneces- sary interruption, and in sufficient quantity always to insure a minimum STATEMENT OF THE PROBLEM 315 of lost time at the machines as well as while in actual transport. The desideratum will be rapidity of movement as well as ample sufficiency of portable storage capacity (the elimination so far as practicable of man-power trucks being taken for granted), so the material can pass along through the several processes of manufacture with least loss of time and the fewest number of handlings. Problem of the Tool Room. — The tool-room problem likewise assumes a place more important than before. It is necessary that the supply of tools be ample, which is to say, much larger than under old conditions; but the system of stocking and distributing must be greatly improved. Red tape will be eliminated so far as possible, and provision made whereby the workman can quickly communicate his tool needs and have them quickly supplied. This may mean electrical communication in connec- tion with mechanical or pneumatic carriers, or perhaps the latter alone. In plants where the highest organization is for any reason impossible, it may mean electrical communication of some simple sort, in connection with boy-transportation of tools. But at any rate it means a change from present methods, as usually found, to others more in harmony with the spirit of accelerated production. Tool Supply and Maintenance. — This phase of the problem (the tool supply) is concerned also with questions affecting the length of time it is expedient to run tools in particular cases before re-grinding, which in turn is related to that of standard speeds and feeds; and may mean the revolutionizing of the system and methods of sharpening tools. This latter evidently will need to be done by inexpensive labor, in grinders designed to operate with precision to give standard shape to cutting edges with minimum skill, and involves the adoption of a complete system of standards and specifications with respect to tool shapes. It likewise involves such storage facilities for tools that they will be least likely to sustain damage (nicked cutting edges, and the like) in the storeroom, and can be dispatched and delivered with promptness. There come in also such matters as more complete standardization and interchangeability of the parts manufactured, where this is not already carried as far as possible; and the maximum use of gages requiring a minimum of time and skill to use, even where jigs can be used most largely. The Array of Problems.— Such an array of intimately related problems, all affecting the most efficient use of the new tools, may well lead to hesi- tation in the adoption of high-speed steel as the standard tool material. It may seem in some cases to involve an entire reorganization and more or less re-equipment of the whole factory, and at the least a wide depar- ture from existing conditions. As long as it will pay increased returns to both factory owner and worker, clearly the new should displace the old. Gradually, of course; for revolutions such as these, as pointed out in the beginning, must take place naturally, else the whole business is like to 316 HIGH-SPEED STEEL s ° 6 "3 9^ as STATEMENT OF THE PROBLEM 317 be put out of joint. And ultimately the change must come, willy- nilly, even to the most conservatively managed shop. So long as com- petition shall be the basis of business, ever growing keener as it must, the law of survival of the fittest will eliminate from that competition any business which neglects to take advantage of every opportunity for increased and more economical production, such as is offered in the metal industries by the extensive use of high-speed steel. Expert Assistance Desirable. — And who is sufficient for all these things? The factory manager as he is, and his force of lieutenants, possibly may not feel equal to such a reorganization, and probably also are in no position to give attention to the working out of all the details. The employment of an expert to take charge of this part of the situa- tion has been already suggested. Experienced engineers stand ready to undertake just such commissions; and it is merely a matter of time and capital to bring any plant to the highest possible state of efficiency. The lack of the latter item, sufficiency of capital, will of course delay the complete carrying out of plans for maximum productive capacity at minimum cost in a good many instances, but should by no means delay the undertaking and its accomplishment as rapidly as circumstances will permit. The Situation Summed Up. — Summing up the situation as it confronts the factory manager, it seems to be about like this: Old tools are to be displaced wherever it can be shown that the time saved in grinding and the tool cost per piece finished, even if there be no gain in rapidity of operation, shows a substantial saving. The new processes of manufacturing composite or compound tools reduce the cost to such an extent that the difference between their first cost even, and that of the old kinds, is not great enough to be seriously considered; and not infrequently actually allow tools to be made cheaper than before. Old machines can be utilized in many cases almost or quite as effi- ciently as new ones, in the general run of manufacturing; and when not, can often be remodeled to a greater or less extent, as the conditions in the case warrant, so as to be moderately well suited to the new require- ments. In many cases the old machines will be unsuited to the heavier duty necessary, and will need to be displaced by new, of types designed with the special conditions in view. Such new machines, in general manufacturing involving the reduplication of parts, will be as simple as possible in construction, with all unnecessary movements and adjust- ments eliminated. For shops producing mainly such things as require or permit the removal of large "weights of metal, machines of extraor- dinary strength and power will be required. The workman himself is an important factor in securing the largest returns from the new tools, and his co-operation is to be secured through a liberal wage system whereby he, as well as his employer, profits; and 318 HIGH-SPEED STEEL his work is to be so adjusted that when working at maximum efficiency his physical strength and endurance are not overtaxed. The tool department will need to be carefully adjusted to the new conditions, the tools themselves being made within the shop or pur- chased outside, as may be most expedient. In the former case it is absolutely essential that there be expert toolmakers who shall be able to turn out tools of the very best quality, and in the latter case it is important that the tools be bought according to specifications fitting them to the special work required of them. The distribution of tools, and their grinding and keeping in condition, will be so organized that the workman himself is relieved of the need for attending to the matter, while promptly supplied as his needs may require. Transportation and storage facilities, especially auxiliary storage for materials in process, must be adapted to the accelerated production, and so co-ordinated as to eliminate all unnecessary handling and all delays occasioning idle machines and workmen. The transportation system will have its units so adapted and of such number that they will in large part serve for the auxiliary storage. Stationary storage, whether for stock or material in process, is to be in such form as to eliminate handling as far as may be, and to facilitate handling where this is necessary. And finally, supervision of the highest order of intelligence, as applied to the special problems involved in the use of the new tools, is indispen- sable. It is necessary to determine beforehand what shall be the time and method of doing given jobs, fixing these elements and the labor cost at the same time, on a rational basis instead of by guess; and likewise to see to it that the conditions laid down are faithfully carried out. Such special supervision may be trained up within the plant itself; but in general it will save time and insure greater efficiency if the reorgani- zation be done through some outside agency, say through industrial engineers thoroughly familiar with the new conditions. CHAPTER XXI. MAKING A BEGINNING. About Tests. — About the first thing considered ordinarily, after it has been decided to use high-speed tools, is the making of tests to determine the best steel for the purpose. Much good money is wasted in this way, and not. a few disappointments grow out of such misdirected zeal. Not that all such tests are useless. They have their place — which, however, is not at the beginning of one's high speed tool experience. There are on the market a great number of the new steels, possibly more than a hundred by this time; and while they vary more or less among themselves as to composition, and therefore as to special adaptation and universality of use, almost any one of them put forth by a manufacturer of repute will do as well as another while a beginning is being made and the local problems investigated. In the meantime the simpler, as well as the better way, is to select some one standard make of steel, and use it exclusively until there shall have been sufficient experience in the making and use of the new tools to permit intelligent experimentation and rational conclusions. Besides the small value to be placed on tests conducted by neophytes, or even by experts, for that matter, under conditions not thoroughly understood, there are several manifest dis- advantages in keeping on hand a supply of each of several kinds of steel, and making or using several kinds of tools for the same work. All these difficulties and disadvantages are avoided if, as suggested, but one make of steel is selected. Obviously the selection must be made with some care, so as to make it reasonably certain that the steel adopted for the time being shall be well adapted to the general run of work done in the shop. It may be mentioned in passing that cheap high-speed steels are to be looked upon with suspicion — at this early stage of experience at any rate. Scope of Profitable Experimentation. — Neither is it worth while to under- take expensive and long drawn out experiments (others would be of small value) to determine cutting angles, tool shapes, standard speeds, and the like data of a general nature, as a basis for the introduction of high- speed tools. It may be possible to improve upon the determinations already made by others, or possibly to modify now accepted conclusions; but experiments looking toward this end may well be left until later in one's experience, and the laws and facts already established and avail- 319 320 HIGH-SPEED STEEL able for reference adopted, for use until mayhap better can be found in the natural course of events. It may be said in passing that to carry on a series of experiments such as these, so the results will have positive value, requires much experience and clear thinking, and involves a preliminary arrangement of conditions not always easy to secure. Thus, to mention a single point, in testing one steel against another, the results will be of little value unless the tests be made at the same time, on the same piece of work or on pieces ascertained beyond doubt to be of exactly the same characteristics as to hardness, etc., with the same feed, speed, and cut, on the same diameter, with tools of precisely the same form and treated so as to develop the maximum possibilities of each — which treatment may perhaps not be exactly the same for both tools. If allowance has to be made for variation in any one of these conditions, the comparative values cannot be established with certainty. The introduction once made and the new methods once fairly established, the workmen edu- cated to the proper management and use of the new tools, there will be time enough to carry out any such comparative tests as may be found desirable. Especially if it is attempted to manufacture all, or perhaps but a few, of the tools, even under the direction of experts familiar with the new steels, there undoubtedly will be enough troubles and problems without the additional distractions incident to such tests. Each Plant a Problem in Itself. — In making a beginning, sweeping changes will be avoided, the development of the problem taking a natural course which will merely modify conditions as fast as expedient, until the whole situation shall have been harmonized with established best practice. Any business organization, particularly that of a big factory, is a delicately balanced mechanism which cannot well be rudely dis- turbed without unlooked-for consequences. In the preceding chapter the sweeping nature of the changes generally requisite for high efficiency have been indicated. However, it is not only inexpedient to make such a change suddenly, but practically impossible. The study of the local problem in any particular plant, the determination of what is best in the case of each of the possibly several thousand operations there performed, will require a long time and much patient study. The obvious thing is to place the undertaking, as already recommended, in charge of a competent man or a group of men selected primarily because of their experience with the new tools and their open-mindedness toward new ideas. The fixing of responsibility in one or more persons is very im- portant. If the matter be left to the several foremen, the work will of necessity be more or less haphazard, there will be lack of uniformity in the method of attack, the experiences of one department will more than likely be lost on another, so that much work will be unnecessarily dupli- cated, and in general the results will be far from what might be attain- MAKING A BEGINNING 321 able with expert supervision and, hearty cooperation from all concerned. With all the other conditions favorable but without such cooperation, the results certainly will still fall far short; and to obtain it is in itself no small problem. Enlisting Cooperation. — The conservatism and self-sufficiency of fore- men is first to be met and overcome. Often this can be done through a sincere endeavor to take them " into the game " and consult with them to the fullest extent. The granting of a bonus for increased efficiency of their departments, as indicated by increased product and lowered cost, is helpful. Above all, frankness with them, and the in- culcation of a spirit of enthusiastic loyalty throughout the previously existing relation between them and the management, will have smoothed the way completely. This is not a treatise on shop management, in the accepted sense; but it seems worth while to remark that the value of personal loyalty to a business, of real interest in it on the part of not only the supervisory force, but of all employees, is a most valuable asset — ■ and unfortunately one to the conservation and development of which little attention seems generally to be paid. The Workman an Important Factor. — While on the whole workmen are somewhat suspicious of efforts on the part of a management to enhance their productive capacity and the individual output, this atti- tude is much less in evidence in a plant where such a policy prevails as that just mentioned — that of frank and square dealing with em- ployees, comprehending among other things a disposition to take the workmen into partnership, so to speak, and give them an opportunity to profit by increased exertion rather than to " cut the whole hog" through a piece-work wage system which fixes a maximum wage for •every job and keeps for the management every advantage arising from increased efficiency and effort. In short, no greater mistake could be made, at the very beginning, than to overlook the workman who is to use the new tools. He must in the nature of the case exert greater activity and work at higher tension, and is justly entitled to a share of the profits If no such spirit of cooperation has previously been culti- vated, the inauguration of a high speed tool regime will be an excellent time and method for introducing also some method of wage payment which shall permit the workman to share in the profits involved in higher efficiency and increased energy. Unless some such provision be made, assuredly the attempt on the part of the management to monop- olize the advantages will properly be resented, and in consequence the returns will be materially cut down below what they might be. Furthermore, to insure permanent good feeling and continued high efficiency, it is necessary that such increment in the workman's reward shall be permanent. The policy in vogue in most piece-work shops of granting a slight increase for greater effort, only to make a reduction 322 HIGH-SPEED STEEL later, when the newness of the thing has worn off, a reduction putting the workman on the same level of reward as before but with harder work to do, is as shortsighted as it is greedy; and it inevitably brings its own punishment — only the management generally is too obtuse to know it, even when it is a continuous occurrence. Losses are none the less large because incurred in ways intangible or unobserved. Tool Problem First Considered. — In a shop where no systematic effort has been made to take large advantage of high-speed tools, it is well to make a start by inquiring into the facilities for tool making and main- tenance (grinding included), and the extent to which tools will have to be bought outside. Naturally this will depend very largely upon the nature of the plant and the number of tools used. If very many, it will pay to organize a tool-making department, or to reorganize an existing one to fit the new conditions, and to secure the service of one or more expert workers with high-speed steel, according to the needs. The equipment necessary, and the methods involved in tool making, are described in the chapters dealing with that subject. In smaller estab- lishments it will be simplest, and probably because it would be inex- pedient to provide a suitable tool-making equipment and to man it with expert service it will be safest also, to buy tools made to specification. Whatever may be decided upon with reference to this point, it will be very necessary to provide adequate and dependable facilities for grind- ing the tools, in order to keep them in proper working condition. Neglect of this point (see chapter on Grinding) will exact the penalty in reduced efficiency and lowered returns. If for any reason hand grinding is neces- sary, in spite of the higher cost compared with machine grinding utiliz- ing relatively unskilled labor, it must be carefully done by skilled hands working as closely as possible by this method to standardized shapes and angles. Capabilities of the Equipment. — Also, whatever the conditions prevail- ing in the shop, it is very important that a careful inquiry be made into the extent and nature of the equipment, and a full report made upon each machine. This should show its kind and type; its capacity; limita- tions as to speed, traverse or feed, and cut; kind of work for which it is adapted and the kind for which it can be adapted; if capable of strengthen- ing or remodeling for higher duty and to what extent and at what probable cost; the kind and probable cost of a machine of maximum capabil- ities to take its place; and such other data as may be found to have a bearing on the problem in hand. With these data in hand the questions as to the routing and placing of jobs under the new conditions will be much simplified, and the purchase of new machines, when this shall become necessary and expedient, can be made, intelligently; without them the re-routing which may be necessary to insure the best results possible without purchasing a new machine, or the purchase of it if that MAKING A BEGINNING 323 be feasible, becomes more or less guesswork. And guesswork, of all things, is studiously to be avoided if results are to count. Ill Advised and Unfortunate Experiences. — A word may not be out of place here in reference to certain ill advised attempts to use high-speed tools and machines. It happens not infrequently that the management of a shop becomes acquainted with the advantages obtainable through these, and, relying upon there presentations and guarantees of steel and machine makers, purchases a supply of steel and installs the machine — only to make a signal failure in respect of realizing expectations. The unfortunate result, in such case, of course is not attributable to either steel or machine, but to their unintelligent use under conditions which would preclude the attainment of satisfactory results. Attacking a Specific Problem. — Assuming that all preliminaries are arranged, so far as can be anticipated, it is in order to take up the con- sideration of particular jobs which may first be changed over. Evidently it will be desirable to take up immediately those cases where the work is especially trying upon the tools in use; and because of the relative simplicity of the tools and conditions, preferably turning operations. Concretely, suppose the turning of a light, high carbon shaft, requiring a long cut, be considered. The first point to determine is that of the tool to use. Only a moderately good finish being required, a standard round- nose tool will fit the case; and since the material is high carbon, the standard tool for this class of work (clearance 6 degrees, back slope 8 degrees, and side slope 14 degrees, making the lip angle 68 degrees — see chapter on Design of Tools), f-inch shank, is selected. The incli- nation to use tool holder stock with a view to economizing in the tool cost is best resisted, though it may possibly be found later that a com- posite tool is permissible. By reference to a table of standard feeds, cuts and speeds, it is seen that with this tool working on this kind of steel at the required depth of cut (for the particular operation here considered, ^ inch), the maximum speed under Taylor standard con- ditions, according as the feed is £%, ^, T \, or -£% inch, is 110, 73.4, 49.3, or 39 feet per minute. By actual trial it is found that a feed of 3 * 2 inch per revolution will leave a finish sufficiently good to pass inspection; and this is therefore selected as the standard feed for the operation. The maximum speed then permissible, if the tool is to last 1J hours per grinding, is 73.4 feet per minute. Summarizing, we have speed, 73.4 feet per minute, feed ^ inch per revolution, depth of cut ff 3 2 inch. Limitations Found and Changes Made. — Consulting the report on the machines employed (two) on the operation, it is seen that these are incapable of running at so high a speed, mainly because they are some- what worn and consume too much power in overcoming the friction of the machine itself when driven so rapidly. A study of the situation and consultation of the reports for other machines shows that an exchange 324 HIGH-SPEED STEEL can be made whereby a pair of other lathes become available without disadvantageously affecting the routing of the piece, which lathes are capable of running at the required speed. This latter, however, is not attainable under the existing conditions. The nearest available speeds are 88 and 60 feet per minute. It then becomes necessary to change the countershaft pulley or the driving cone, or both, to get the required speed; and in making the change the belt and the belt faces are widened TOOL TEST NO. NO.- OPERATION- PIECE MATERIAL- _ MACHINES OK SPINDLES OPERATED " CONDITION- LUBRICANT, Record of Grinds No. Mill. per Grind Pieces Finish d No. Min. per Grind Pieces Finish'd No. Min. per Grind Pieces Finish'd 1 16 31 2 17 32 3 18 33 4 1!) 34 5 20 35 6 21 36 7 22 37 8 23 38 i) 24 39 10 25 40 11 26 41 12 27 42 13 28 * 43 14 29 44 15 30 45 Total Total Total Production Record Day Finish'd Scrap'd Day Finish'd Scrap'd Day Finish'd Scrap'd Total Date,_ Foreman Department . . Fig. 269. Tool test record form. Filled in mostly by the workman, or by the one conducting the test. It is convenient to indicate different tool steels by differently colored cards. Comments made on back of card. so as to give a sufficient margin over what would be required to pull the expected load. Result, a peripheral speed of 71 feet for the shaft to be turned. The feed mechanism is found sufficiently strong for the work required. The tail center is replaced with one of high-speed steel, and the bent-tail dogs previously used, with a quick-acting chuck. And the running test is begun. Practicable Conditions Established.— It transpires that a back rest is required, and that the speed 71 feet per minute, in spite of the rest and MAKING A BEGINNING 325 of the preliminary report, is so high for these machines that the vibra- tion set up does not permit sufficient accuracy in the work and also effects a reduction in the time after which the tool must be ground; hence it becomes necessary to reduce the speed or to install new ma- chines. The latter course being for the present inexpedient, the cutting TOOL TEST NO jeiECE OPERATION-. MATERIAL CONDITION T.OOL__ „ SIZE ANNUAL REQUIREMENT _ PIECES Tool Number No No Gain Speed, (Feet per minute ) Eeed, (Inches per Revolution) Cut. (Depth) {ipm'aTell °P erated Cutting Timej (per Piece) Grinding Time, •• <• Time Allowance " ■ Actual Time for Operation, Daily Output Rate per Hundred Pieces Cost of Tool " " Cost of Power « » Interest & Depreciation Hundred Pieces Overhead Expense " >« Cost of Scrap " " Total " " Saving on Year's Requirement . ( Machines > , } A ' Spindles > Daily Wages of Workman — Former,. Days ;New, Duration of Test, New Rate in Effect, Date._^ ..Department Fig. 270. Record of data and results. Cost of power, depreciation and interest, and overhead expense are rarely of sufficient importance to require consideration. Remarks made on back of card. speed is reduced to 58 feet per minute (provision having previously been made for such a contingency) and the result is found to be all that could be desired in respect of finish, while the tool endurance rather more than equals that anticipated. It remains then to calculate the economy effected and to rearrange the labor cost accordingly — not forgetting to take the workman into account, as already pointed out. Data Determined. — From the factory records it is found that the yearly requirement is 24,000 pieces; average daily output (two lathes), 92 pieces; piecework rate, $2,125 per hundred; average daily wage for 326 HIGH-SPEED STEEL workman, $1.91. Also, determined either from records or observation with stop watch, as the case may be: actual cutting time, per piece, 8 minutes; time allowance for grinding and setting tool, average, 1 min- ute per piece; time allowance for handling piece, cleaning machine, and other losses, 4 minutes: or a total time allowance of 5 minutes, and a total operation time of 13 minutes per piece. The actual operation INSTRUCTION CARD. Piece No.. Operation - -Department . -Order No Fixtures ) Jigs J Finish Gauge -Machines Tool to use- Cutting speed- Revolution of Table feed : Traverse tool -feet per minute per minute -feet per minute = -per stroke revolution Depth of cut- Change tool every- -minutes Lubricant or cooling agent- Expected daily output- Directions: Date- -Signed- Fig. 271. Card of instructions sent to workman with new tool when job is changed over. time is half this, or 6J minutes, since two lathes are used and two pieces are finished during each period of 13 minutes. The data observed (or computed beforehand, if the supervisory force has sufficient skill and experience to determine this without recourse to MAKING A BEGINNING 327 experiment) bring out that the actual cutting time under the new con- ditions is 4.5 minutes; time for changing tools, etc., 0.2 minutes; time for handling and changing piece, 2.5 minutes; or a total time of 7.2 minutes per piece. But since two lathes are used, the real time per piece becomes one half this, or 3.6 minutes; and the daily (10 hour) output is about 160 pieces, or an increase of rather more than 75 per cent. On a piecework basis, allowing the workman a substantial increase in pay (suppose we say a day rate approximating $2.30, as against his former $1.91) because of this greater exertion, the cost of the job can well be reduced to $1.40 per hundred, effecting a saving $0,725 per hundred pieces, which in a day amounts to nearly $1.50, or an economy of more than $220 on the year's requirement. Under a good premium or bonus wage system even this showing would be bettered. The Element of Tool Cost. — This item, however, is by no means the only one through which economies are effected. There is, for example, the matter of tool cost. In most cases it might perhaps be expected that this item would be increased. As a matter of fact it rarely is so, because of the greater life of a high-speed tool and the consequent distribution of its first cost over a greatly increased output. In this particular job it turns out that there is a saving in tool cost approximating $0.08 per hundred pieces, or not far from $20.00 for the year's requirement. This does not include tool maintenance and grinding, in which item there would be another considerable saving, since this job was very hard upon carbon and self-hardening tools. There is also the matter of the largely increased time during which these standard machines become available for other work. Whereas before the change two lathes were required practically the whole working year (about 260 days) to get out the re- quirement, under the new conditions they are required for a maximum of only 150 days, leaving them available for about half the time for other work and consequently reducing capital and maintenance account by nearly one half. Saving in Scrap. — Furthermore, in this job the scrap loss was pre- viously an important factor — almost half as great, indeed, as the labor cost; and this is reduced, if not to a negligible item, at any rate to a reasonable minimum for the job. The saving through this item is shown, along with others, in the following table. Data Summarized. — Quite evidently this is not a typical factory job. Nevertheless it is representative of a distinct class of operations to be found in most plants; and it has been selected as an example in order to indicate as clearly as possible the factors to be considered in changing over jobs from ordinary to high-speed tools. About the same points are involved in almost any other operation of the kind with which we are at present concerned, and the method of attack will be about the same. 328 HIGH-SPEED STEEL TABLE X. Item (per 100 Pieces). Self-hard- ening Tool. High-Speed Tool and Equipment. Economy Ef- fected . Total, on Year's Re- quirement. Piecework rate $2,125 0.089 0.072 1.050 $1,400 0.004 0.041 0.160 $0,725 0.085 0.031 0.890 $174.00 20.40 7.44 213.60 Tool cost (including sharp 'ng) Insurance and interest, equip't Scrap, net loss Totals $3,336 $1,605 $1,731 $415.44 1 Neglected. Classification of Jobs. — So-called " try-outs " are useful in connection with the first introduction of high-speed tools, and perhaps are neces- sary; though the method, the time, and the order of doing the work can almost, if not quite as well, be determined in the office and prescribed for the workman without actual test, where a proper knowledge of the conditions exist. In fact it would be quite out of the question to under- take a test or to make a try-out of every one of the tens of thousands of operations in a big plant. It is enough that this be done, if at all, in a relatively few cases which are selected as typical of most those met with. All jobs are then classified according to their characteristic features, and standards established for the several classes. Evidently the analysis of a great number of very different jobs will present manifold difficulties, and some will seem to defy classification. It is imperative, however, that this be done so far as possible in order to minimize the time required for making changes and calculations. There must of course be some rational basis for such a grouping of jobs, though this may vary more or less according to local conditions and the specific nature of most of the operations. The factors here indicated are of sufficient importance to require consideration, and may be taken tentatively, at least, as a basis : a |TyP e °f machine on which job is done. [Limitations and capabilities of machine available. (Material worked upon, and its particular qualities. Shape and other special characteristics of the piece operated upon, including size. Kind of tool to be used, and its capabilities. „ Amount of material to be removed. ' Possibility of using multiple tools. Feasibility of lubricating or cooling tool or work. D. Finish required. MAKING A BEGINNING 329 Other elements of course enter into many jobs; very likely in not a few instances they may assume importance beyond some or perhaps even all those here pointed out. This, however, is unlikely to be so, and these may safely be taken as fundamental to the determination of standards for operations except as special cases may arise. The standards once established, it should be possible to fit any given job nearly enough to a pretty clearly defined class, and to modify the conditions later as occasion or experience may indicate. APPENDIX A. ANALYSES OF HIGH-SPEED AND SPECIAL STEELS OF VARIOUS MAKES. 02 S 3 ■3 03 C > B 3 C 0) .a O i=5 a c 3 H a 3 a o 5 c o .o u 03 0) o a oj bo c 03 <=; c o ._o in W o o 3 - 02 1 la 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 26a 26b 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 0.32 0.29 17.81 18.19 16.19 14.41 17.61 14.23 25.45 14.91 17.79 19.64 18.99 23.28 18.93 13.44 24.64 19.97 19.16 9.25 16.00 16.00 14.71 15.31 14.91 14.62 5.95 5.47 3.86 3.28 4.24 3.44 2.23 5.71 2.84 2.85 2.61 2.80 3.52 3.04 7.02 3.88 5.61 6.11 3.50 3.50 2.90 2.88 2.80 2.81 4.30 3.67 2.95 2.85 2.93 2.70 4.10 2.69 2.90 5.11 6.00 4.58 0.682 0.674 0.736 0.709 0.502 0.739 0.838 0.790 0.650 0.760 0.670 0.800 0.580 0.760 0.600 1.280 0.790 0.320 0.700 0.700 0.700 0.540 0.450 0.600 0.900 1.160 0.705 0.791 0.800 0.250 0.370 0.640 0.750 0.490 0.650 0.560 0.650 0.620 0.440 0.550 0.550 0.660 0.790 0.660 0.320 0.400 0.540 0.500 0.940 1.030 1.190 1.250 0.07 0.11 0.06 0.07 0.10 0.06 0.29 0.06 0.12 0.30 0.20 0.11 0.19 0.09 0.03 0.14 0.049 0.043 0.210 0.120 0.240 0.165 0.034 0.060 0.087 0.090 0.265 0.165 0.125 0.052 0.205 0.220 0.48 0.013 0.012 0.014 0.015 0.029 0.009 0.009 0.016 2.03 4.21 0.28 7.60 0.13 0.081 0.12 0.12 0.10 0.18 0.12 0.10 0.01 0.196 0.133 0.090 0.320 0.481 1.340 0.017 0.018 0.018 0.017 0.016 0.024 0.013 0.010 0.009 0.008 0.009 0.008 0.008 0.008 0.75 6.25 10.68 14.91 14.29 13.40 17.27 16.48 18.66 14.83 17.60 19.00 16.75 0.06 0.18 0.24 0.08 0.020 0.035 0.014 0.012 0.020 0.010 0.018 0.020 0.016 0.020 0.017 0.016 0.008 0.179 0.090 0.393 0.008 0.014 0.010 0.007 0.019 0.007 0.006 0.010 0.010 0.010 5.19 0.10 0.17 0.20 0.200 0.110 0.038 0.78 9.61 13.00 17.54 19.09 17.81 19.03 14.27 18.40 15.29 16.22 10.08 13.76 4.78 0.46 7.56 2.25 2.88 2.72 2.69 2.48 3.55 2.89 1.85 4.73 3.00 4.49 0.69 0.00 3.34 0.28 0.12 0.27 0.11 0.08 0.12 0.15 0.18 0.16 0.07 0.27 0.30 0.46 0.85 0.100 0.070 0.090 0.036 0.150 0.135 0.340 0.120 0.015 0.028 0.010 0.019 0.025 0.008 0.008 0.006 0.008 4.38 0.34 0.210 0.110 0.118 0.200 0.210 0.015 0.010 0.025 0.024 0.005 0.010 0.009 0.025 330 APPENDIX 331 MEMORANDA. Analyses numbered from 1 to 25 inclusive are those given by Taylor. 1 and la are the same steel, the one referred to by him as the best of all those used in the Taylor- White experiments. 44 is the average of the analyses of 12 different melts of the same steel, or rather of steel marketed under the same name. 45 is high-speed, but not of the highest grade. 46, 47 and 48 are steels recently put upon the market as " semi- high-speed " or " intermediate " steels. 49 is sold as a steel especially adapted to finishing cuts. It is not at all in the high-speed class. APPENDIX B. FRED W. TAYLOR ON THE VARIOUS METHODS OF HARD- ENING HIGH SPEED STEEL TOOLS. ' Special Treatments Unnecessary. — For some years past it has been rather amusing to us to hear the special directions given by the various manufacturers of steel suitable in chemical composition for making the high-speed tools. Very frequently a tool steel maker implies, or directly states, that the chemical composition of his particular high speed tool steel requires " special treatment." The fact is, however, that our recent experiments demonstrate beyond question the fact that no other method which has come to our attention produces a tool superior in red hardness {i.e., high speed cutting ability), or equal in uniformity to the method described. This applies to all makes of high speed tool steels which are capable of making first-class tools, whatever their chemical composition. Various Methods Tried. — It is the writer's belief that during our long series of experiments at the Bethlehem Steel Company, in our search for uniform tools and for the method of imparting the highest degree of red hardness to tools, we tried substantially every method which has since come to our attention. For instance, in giving the tools the high heat we heated them in a blacksmith's coke fire, a blacksmith's soft-coal fire, in muffles over a blacksmith's fire, and in gas-heated muffles. We also constructed various furnaces for this purpose. We heated tools by means of an electric current, with noses under water, and out of water, and by im- mersion in molten cast iron. Moreover, by every one of these methods we were able to produce a first-class tool, provided only the tool was heated close to the melting point. Cooling Experiments. — In cooling from the high heat we experi- mented with a large variety of methods. After being heated close to the melting point, tools were immediately buried in lime, in powdered charcoal, and in a mixture of lime and powdered charcoal; thus they 1 The extensive investigations, and prodigious amount of time and labor devoted to them in the development of high-speed steels and their treatment, give to the follow- ing extract from Mr. Taylor's address or report a special significance. The para- graphs are 1001 to 1006 inclusive, at pages 200 and 201 of the address "The Art of Cutting Metals," already mentioned. 332 APPENDIX 333 were cooled extremely slowly, hours being required for them to get below a red heat. And we wish clearly to state the fact that tools cooled even as slowly as this, while "they were in many cases quite soft and could be filed readily, nevertheless maintained the property of " red hardness " in as high a degree as the very best tools, and were capable of cutting the medium and softer steels at as high cutting speeds as the best tools which were cooled more rapidly and which were much harder in the ordinary sense. Tools were also cooled from the high heat in a muffle or slow cooling furnace with a similar result. On the other hand, we made excellent high-speed tools by plunging them directly into cold water from the high heat, and allowing them to become as cold as the water before re- moving them. Between these two extremes of slow and fast cooling; cooling in lime, charcoal, or a muffle, on the one hand, and in cold water on the other; other cooling experiments covering a wide range were conducted. We tried cooling them partly in water and then slowly for the rest of the time; partly in oil, and then slowly for the rest of the time; partly by a heavy blast of air from an ordinary blower and the rest of the time slowly; partly under a blast of compressed air and then slowly. We also reversed these operations by cooling first slowly and then fast, as described. We also cooled them entirely in an air blast and entirely in oil, and then partly first in oil, afterward in water, and then first in water and afterward in oil. Good Tools by all Methods. — By every one of these methods we were able to make a good high-speed tool; i.e., a tool having a large degree of red hardness, and capable of cutting at very high cutting speeds. But by none of these processes were we able to obtain tools as uniform and regular as those produced by our lead bath and air cooling. APPENDIX C. REFERENCE TABLE FOR DETERMINING CUTTING SPEEDS. Feet per Minute. 5 10 15 20 25 30 35 40 45 50 Diam. in Inches. REVOLUTIONS PER MINUTE. i 2 38.2 76.4 114.6 152.9 191.1 229.3 267.5 305.7 344.0 382.2 1 30.6 61.2 91.8 122.5 153.1 183.7 214.3 244.9 275.5 306.1 f 25.4 50.8 76.3 101.7 127.1 152.5 178.0 203.4 228.8 254.2 i 21.8 43.6 65.5 87.3 109.1 130.9 152.7 174.5 196.3 218.9 19.1 38.2 57.3 76.4 95.5 114.6 133.8 152.9 172.0 191.1 n 17.0 34.0 51.0 68.0 85.0 102.0 119.0 136.0 153.0 170.0 15.3 30.6 45.8 61.2 76.3 91.8 106.9 122.5 137.4 153.1 if 13.9 27.8 41.7 55.6 69.5 83.3 97.2 111.1 125.0 138.9 ii 12.7 25.4 38.2 50.8 63.7 76.3 89.2 101.7 114.6 127.1 if 11.8 23.5 35.0 47.0 58.9 70.5 82.2 93.9 105.7 117.4 if 10.9 21.8 32.7 43.6 54.5 65.5 76.4 87.3 98.2 109.1 if 10.2 20.4 30.6 40.7 50.9 61.1 71.3 81.5 91.9 101.9 2 9.6 19.1 28.7 38.2 47.8 57.3 66.9 76.4 86.5 95.5 21 8.5 17.0 25.4 34.0 42.4 51.0 59.4 68.0 76.2 85.0 21 7.6 15.3 22.9 30.6 38.2 45.8 53.5 61.2 68.8 76.3 2f 6.9 13.9 20.8 27.8 34.7 41.7 48.6 55.6 62.5 69.5 3 6.4 12.7 19.1 25.5 31.8 38.2 44.6 51.0 57.3 63.7 31 5.5 10.9 16.4 21.8 27.3 32.7 38.2 43.6 49.1 54.5 4 4.8 9.6 14.3 19.1 23.9 28.7 33.4 38.2 43.0 47.8 41 4.2 8.5 12.7 16.9 21.2 25.4 29.6 34.0 3?.l 42.4 5 3.8 7.6 11.5 15.3 19.1 22.9 26.7 30.6 34.4 38.2 51 3.5 6.9 10.4 13.9 17.4 20.8 24.3 27.8 31.3 34.7 6 3.2 6.4 9.6 12.7 15.9 19.1 22.3 25.5 28.7 31.8 7 2.7 5.5 8.1 10.9 13.6 16.4 19.1 21.8 24.6 27.3 8 2.4 4.8 7.2 9.6 11.9 14.3 16.7 19.1 21.1 23.9 9 2.1 4.2 6.4 8.5 10.6 12.7 14.9 17.0 19.1 21.2 10 1.9 3.8 5.7 7.6 9.6 11.5 13.4 15.3 17.2 19.1 11 1.7 3.5 5.2 6.9 8.7 10.4 12.2 13.9 15.6 17.4 12 1.6 3.2 4.8 6.4 8.0 9.6 11.1 12.7 14.3 15.9 13 1.5 2.9 4.4 5.9 7.3 8.8 10.3 11.8 13.2 14.7 14 1.4 2.7 4.1 5.5 6.8 8.1 9.6 10.9 12.3 13.6 15 1.3 2.5 3.8 5.1 6.4 7.6 8.9 10.2 11.5 12.7 16 1.2 2.4 3.6 4.8 6.0 7.2 8.4 9.6 10.7 11.9 17 1.1 2.2 3.4 4.6 5.6 6.7 7.9 9.0 10.1 11.2 18 1.1 2.1 3.2 4.2 5.3 6.4 7.4 8.5 9.6 10.6 19 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.1 10.1 20 1.0 1.9 2.9 3.8 4.8 5.7 6.7 7.6 8.6 9.6 21 .9 1.8 2.7 3.6 4.5 5.5 6.4 7.3 8.1 9.1 22 .9 1.7 2.6 3.5 4.3 5.2 6.1 6.9 7.8 8.7 23 .8 1.7 2.5 3.3 4.1 5.0 5.8 6.6 • 7.5 8.3 24 .8 1.6 2.4 3.2 4.0 4.8 5.6 6.4 7.2 8.0 25 .8 1.5 2.3 3.1 3.8 4.6 5.3 6.1 6.9 7.6 26 .7 1.5 2.2 2.9 3.7 4.4 5.1 5.9 6.6 7.3 27 .7 1.4 2.1 2.8 3.5 4.2 5.0 5.7 6.4 7.1 28 .7 1.4 2.0 2.7 3.4 4.1 4.8 5.5 6.1 6.8 29 .7 1.3 2.0 2.6 3.3 4.0 4.6 5.3 5.9 6.6 30 .6 1.3 1.9 2.5 3.2 3.8 4.5 5.1 5.7 6.4 334 APPENDIX 335 The revolutions per minute for any greater speed may be obtained by multiplication and addition. Suppose it is desired to find the r.p.m. of a milling cutter having a diameter of \\ inches and expected to run at a peripheral speed of 85 feet per minute. The required number may be found by following the line opposite the given diameter, 4J, to the column under 40, where is found the number 34.0; and also to the column under 45, where is found the number 38.1. Adding these numbers we have the required r.p.m. for a speed of 40 -f 45, or 85 feet per minute, which is 72.1. If the r.p.m. for a surface speed of, say 120 feet, is re- quired, it may be found by multiplying the number in the column under 20 by 6, which in the case of a 2f inch cutter would be 27.8 X 6 or 166.8. Conversely, the surface speed required for a given diameter and r.p.m. can be determined, though for this the use of the following formula is simpler: S = D X R X .2618, in which S is the surface speed in feet per minute, D the diameter, and R the revolutions per minute. We also have: R X .2618 S 336 APPENDIX APPENDIX D. © © o © °° e 5 o p e o o ss o n e I 1 J3fl o 3JT10 o IT p o >B O cs 00 t- s s 10 « t- X X ■~T"' x i x \ x \~ X - f - --- x. ° :s: ... . ■ \ \ i p ■ \ k . x x -x"-" , . x X X _L °° .x,_ X \ ■^"^ .. , -■'- X. ~^x" j ■* •^T-r \ _ _ *- V x V \ \..x X S| \ ■ \ V x : x X x_ x \""* x X^ - -, i>! _s_ is x xk ■ x X - " ^~ x s x x ,i^i " o \ \ X \ * \ ■ xh I X x X \ X V \ \ \ x 'x'x x . ■ "\ \ \ x- X X X (X X x V X x x ^, X \ X X 4-x - X X. • x - x s ^^ x X--" "".^S ' I ' X \ X x. s s s X \ K :s_ s x x~\ X _ ■x s^v ~ x:!:; x V J? X x^s s ' x_ x,S X s X ~ . x. x x i s ^x x X :s x x . x>_v ! \x..X r i_\. X x i.s: X x s -""\! .N 1 O . x- '" X xl \x\ x ^\ x ■> X ^ X X^ x X \ r — =«o v \ s x^ W X \ T . XX x^ X \x\ x xx s X \ x "^ c ' .\i \\ \ A 's s > \ ■> x- \ N xN\\ \\\ \ X \ \ 'x \ X X \ X x X X ■ — «- >- \ X \ \ \- x\ \ x \ xf\ \ x \x\x xXX X \ ^x X s X \ x X x ■ - ^ ~ v ^J^ :: 3:?: S"-x as ^X^ x X, x x X x X X > -s: :>?£ XX X K x ^ ^ 2S^xx -'x & rs X X X x x X X X \ X • -\s V;- '. ' 5 X X * \ XXXX- X* X 5 X X X x x X X X X \ X £ " " • X x- X x X X X , - X \\x. xx X \ X\ W X x \ x x X .-X _, x, \W x- X ..x v > \ \ lx x ^ : X -x ph \ s V- ^ X x s \ . X X -x \ X . OS ", \ 'xS xx X l -x x X \\ S X.-&V-- x.\ s V\ s ^5 X X '" X X S~ 5, ";-■* - l v\ X J X V X x \4 ^ X s xS.x\ S^iX. ' S X \ X -v S \ — _v. ■ x - X x-S v^s T _>T - V gj -.. " — x-.ir- 1 -V. 3 X x s~ xx X ■x x h X C s .i X x \ X x, v x ^ x \^ X X xK > X X r^ \\ X X x, X 5 X NX x X x ;\\ X S V S^x - " X X s X x \x^^ s S -V x x X X x OS ; pi ->,S v2 x x ' x X \ s s , x 5 X x ^VX, X x X x X X X x x— g x\ X X \ X X xT \ Nxx xX X \ ^x \x v • x \\ \ \ XX S Xx- * ^ r_ \ V s \x i ft [ X x S, V x\xx, xXX.x - x l\ \\x \ s v x s ■v x x x \S s \ s ■xx i-- s " o xx- ^ ^ w \> - b X \ \ \ <> ,\x\X \\\x ,- K W s\\ \ „\ x'x x . X >3 XX \x\ X ■* „ X N ^5 \% \ \ s \ X wwl x\xx^ v\ \ v- x-\ \ ^.Xj.i x \ \ X M CT -^ X i\ * X ■ X x ^ 5;-' x:\ x , M L Sx , "■ cs ■^^s XX? pr b* .a »!..§> x X X x x X Ns \x, \ 5 ,> V X N N H \ v . , x X « . XX b^? '. x \ x x| \ s: \AV \S S%- X SX x \\X X x ;. xr ■a x - x ^Sx XX X X V X, x w « i ^N,x tffc \ x- •X X V 1 X x \ X \ XX \ ': \'x\ X X N X, l x\ x x N X X x . ~ <» & 1 K.X XX x x. X x X x x.X x- ;x X xx. i^^ X x xx. X \ w V \ x X x x X XXXX, xx-x X x \ ^' s \ X x. >." 1C ^-K^x . x \ X x-^ " XX x.X - X x > X x ^ '^; ■>', X-X xl X ^^x^ X x^ x x; \X X !5 \- r : ^ .xi X - ' •■ , - x X V - x X x XXX -\ x\ ^ X X x x x ~X» xV x ' \ x X XX. X . X \ ^ K X XX s x x X X^ \ . T- \ - x ^ X^x sV S s\ ■<\ s XX \- x\N^ xX X V, xN X OS ~ \ v\ \- x s x\ |x w X X - x X'x ' x \- xN Nx\ xx-, ^ x \ X 00 -R "x" \ x x \' s X x\ ixx ^ X x xx'- xW sS N N>» ^> \ ? ; x\Nx X X s\ x:i_t_ J* X X V V X -txx | vXX \ \ S ^ xW < \ \>0 -N s \ ^s^ xxxx H^ X a = » N ■x \ 1 \ S X X x ! v l^x x X VX^ "xV \ x \ \ ^>^ ss \ N^ vxx> >x^ V \ \i x * x' -^ = \ X X X, \ \ X X X \ * ; x^x] Nxx^ x^> \ \^ \ \ ^^ x"§ XX xN § X * x._^< \ X X ^ \ \ \ X x x X X \ s[x xTx x x X x ^Cx'x xXXsX :^ \ X ^ ; :i §$ \ \ \ xN ;;; 1 \ ^ ^1 X-ol X 5— w X \ N » X \ \ \ X X X 1 x \ x \ s\ X V S \ X x x^ ^x^\ \ \ s X x^ X \ \ x x^ x xlx X Cx, O o ™ s c i 8 c o o § s o c 5 3 o o o o s 1-1 CO t^ s o , a^uaicu J9d iiaaj n.; peadg APPENDIX 337 As shown in the enlarged section, follow the vertical line representing the given diameter to its intersection with the horizontal line correspond- ing to the required surface speed; follow the diagonal nearest this point 3 Diameter 4 w „1 min. 35 sec. to turn one inch Fig. 273. Enlarged section of Fig. 272, showing method of determining time of operation. up or down, until it meets the horizontal line corresponding to the desired feed as read from the left of the table. A vertical line dropped from this point to the bottom line will there show the time in minutes it will take to turn one inch under the given conditions. Diagram by A. Thompson, re-published by courtesy of Machinery, New York. S38 APPENDIX M I— I On Ph < >1 O w o H o o C Q H !> O o 02 CD »c Oi 05 c3 00 00 1-1 O IO CD >> Q O OS CD t^ co t^ p. •a . +3 >3S 10 O 10 IO o O E" 1 02 10 - 1 - > Wt.of Metal Re- moved. 00 CN (M