Proceedings

of the Tomsk Technological Institute

of Emperor Nicholas II

Vol. 10. 1908. No. 2.

I.

P. A. Kozmin.

The Wheel Lathe and Its Evolution into Modern Types.

With an appendix of XI plates of drawings. 1–52.

The Wheel Lathe

and Its Evolution into Modern Types.

I.

Before proceeding to develop the question proposed, I consider it not superfluous to set forth the considerations that led me to undertake this work.

If we cast even a cursory glance at the development of our railway system and compare the conditions and results of railway operation in Russia with those in Western Europe, and especially in America (the United States), what strikes us most of all is the high cost of operation in Russia, all other conditions being equal, or even more favorable.

For greater clarity I shall draw numerical parallels between our railways and the network of the North American United States, since the railways of that country, by the nature of their traffic, are close to ours, serving a territory chiefly with semi-finished goods and raw materials (pig iron, coal, cotton, grain, timber, and so forth).

Turning to freight transport per 1 pood-verst, we see that carriage per this unit is 26.6% dearer with us than in the United States. This figure is still more striking if one takes into account the fact that the average wage of a salaried employee or worker in the United States is three times higher than with us.

It is clear that only with a high technical organization of railway operation can the railways of the United States yield a large profit at a lower tariff rate, despite high wages.

In illustrating with statistical data the high level of the technical organization of railway operation in the North American United States, one must point first of all to the number of workers and employees per 1 verst: with us, 12.3 persons, while in the United States, 3.5 persons, i.e. three times fewer than with us.

Further, with us there are 31 locomotives per 100 versts of railway, while in America only 14; freight cars per the same 100 versts with us—

719, while in the United States it is 554, with the same quantity of goods carried per unit of track and per wagon axle.

All these data refer to average conclusions drawn from the operation of railways over the last ten years.

But if we know that the budget of railway operating expenditure consists to a considerable degree of expenditure on the repair and renewal of rolling stock, then the conclusion is clear: the reduction of the absolute amount of this expenditure also depends on a more rational equipment of the workshops.

There is no doubt that, with the so-called “American” running of locomotives, the wear of a locomotive in America is much higher than among us, owing to the fact that the mileage of American locomotives is twice the mileage of ours. Likewise, the wagons, in one and the same unit of time, are in service considerably more than among us.

From this it is already quite understandable that the repair turnover of rolling stock in America is much faster than among us, and consequently the workshops there are adapted to a productivity at least twice as great as ours.

However tempting the topic of developing a standard type of American workshops may be, I take upon myself the incomparably more modest task of indicating a standard type only for the wheel lathe, as the heaviest and most expensive machine. And I think that only by means of a series of preliminary, painstaking investigations of individual types of machine tools shall we be able to approach the solution of the problem of rational equipment not only for railway workshops, but also for workshops in general for the mechanical working of metals, approaching the solution of this problem by determining the greatest productivity in connection with the suitability of the designs.

However, alongside the modest task of indicating a standard type of wheel lathe, I shall attempt in this article to define the foundations of those scientific requirements which must be adopted in characterizing and evaluating one or another type of machine for the mechanical working of material in general, and of metals in particular.

It must be regarded as indisputable that the piling up of drawings and descriptions of machine designs in books on mechanical technology, which are supposed to constitute scientific literature for technicians, is merely a poor technical chronicle. A book can satisfy scientific requirements only when, first, it gives a clear idea of the production process, without confusing the essence of the process with a recipe; and, second, when it schematizes the designs of machines and the parts of machines that perform one or another stage of the process. The second requirement is important in that it obliges the author to avoid the accumulation of investigations or manuals

by constructive descriptions, and, most importantly, gives us the possibility of clearly defining the essence of the design and of reading the drawing without lengthy descriptions. By satisfying the first and second requirements, we thus create a criterion for a clear, comprehensive critical appraisal of the machine.

Thus, the process of machining and the scheme of the machine characterizing one or another moment of the process—this is the foundation of every mechanical technology. Needless to say, illustrations of schemes by means of a constructive drawing are admissible and even necessary here. But when the matter concerns a monograph on some definite type of machine having a special purpose, then, besides the working process and the scheme of this machine, it is necessary to unfold before the reader the complete picture of its historical development down to our own day, in order to set the designer’s thought upon the iron rails of logic, which determines the lawful movement of creative thought and removes its fluttering into the realm of fantasy.

Having clarified for himself the working process of the given machine and having become acquainted with the history of the realization of this process by definite organs of the mechanism, the designer who creates a new machine, or the engineer who has to deal with a finished machine, can quite orient himself in this question. His answers will be a logical conclusion from, or continuation of, previous work, if he is a designer; and the pure source of his requirements will be the same logic of expediency, if he is a practicing engineer. And only in this way can we bring the work of creative thought to high productivity and keep it on the straight path of logic, instead of allowing it, under the name of the “genius” of inventiveness, to wander, like the blind, along the paths of fantasy and the creativity of guesswork. That inventiveness is a science and not an art is shown to us by the classical land of inventions and inventors—America, which long ago already took the strictly scientific path, verifying experimentally only the theoretically created construction of the inventor.

*
* *

One of the few machine tools whose productivity can be reckoned with more or less accurately even in the rough practical sense of the word is the wheel lathe. There are no railway workshops for medium repairs of locomotives that could dispense with a wheel lathe—this in the first place; and in the second, the high cost of this machine, scarcely not the most expensive of all the equipment, as well as the highest remuneration of the worker’s labor on it, deserves that we dwell upon this machine and give a critical appraisal of its co—

temporary design. But, as has already been said, a more or less correct assessment of the modern type of this lathe is possible only when we trace the evolution of its design, i.e., become acquainted with the lathe in its historical perspective.

The wheel lathe, both in its historical past and at the present time, is used almost exclusively for turning the tread surface and the side surfaces of the tires of wheelsets. And if formerly the lathe was designed with a device for increasing the speed of rotation for the purpose of turning the axle journals of wheelsets, then the regulation of speed for this same purpose in modern designs of the lathe must be recognized as a constructive atavism of the inventors, as I shall try to prove below.

How great the role of the wheel lathe is in railway work may be judged at least from the fact that, in addition to turning new tires, the re-turning of worn ones is determined by a colossal figure, since the total run of locomotive-tender and car axles annually reaches thirty billion axle-versts. And this, with a mileage between re-turnings of from 15 to 50 thousand versts per axle, creates an enormous number of wheelsets requiring re-turning.

In characterizing the wheel lathe both in its constructive evolution and in its modern, more or less established type, I shall adhere to the following three principles, which I shall also place at the basis of the evaluation of the lathe in terms of its constructive expediency and simplicity:

1. The method of imparting rotational motion to the wheelset being turned and the regulation of this motion.
2. The character of the working tool used for turning.
3. The mounting (support) and movement of the working tool along the machining line.

In the order of these three basic propositions, arising from the very process of the work—removing chips from the tread surface of the tire in order to give it the necessary profile—I shall set forth the present monograph.

The very name of the wheel lathe already shows that the inventor’s first thought must have been to transfer the idea of an ordinary lathe into the construction of a wheel lathe. And indeed, probably, one of the first was designed and built as a lathe schematically represented in Fig. 1. I say “probably” because in the literature for almost sixty years I have found no traces of this lathe, whereas lathes of this design (3 examples) are still operating at the Putilov Works, the oldest of the Russian works (founded in 1801),

and therefore representing a good museum of historical and modern technology.

As can be seen from the diagram, this machine is an exact copy, on an enlarged scale, of an ordinary lathe. In it, the front headstock A with the faceplate Z is fixed immovably. The rear supporting headstock B is movable and, in the construction of the spindle C, represents the supporting headstock of old lathes. The faceplate Z with a toothed rim receives its rotary motion from the shaft V by means of the gear Z₁; the wheelset P is placed on the supporting centers and secured by drivers. Thus, the supporting center of the front headstock rotates together with the faceplate and the wheelset, while the center of the rear headstock is stationary. The axes of the headstocks are mounted on the bed T.

Fig. 1.

The machine is, from the structural point of view, unusually simple, all the more so because it has no set of change gears for altering speeds; but the direct transfer of the idea of the lathe to the wheel lathe is clearly inexpedient. First, this machine, taking no less than 3 HP, strongly wears, by the supporting center of the movable headstock, the corresponding recess in the axle of the wheelset, thereby causing decentering, which adversely affects the regularity of the turning of the wheelset; secondly, the entire working effort is transmitted to one lower gear of the driving shaft, and this causes rapid wear of the gear and wear of the bearing.

Both phenomena are observed when working on these machines, and therefore the Putilov Works, though still not having the courage to discard completely these archaeological adornments of the car-repair machine shops, decided, after numerous repairs, to rebuild one of the most ruined machines and bring it closer to the modern type.

Thus, this design of machine, although simple, is not expedient. The shortcomings indicated above immediately lead the technician to the correct idea of the construction of a wheel lathe. It is clear that the machine

must be constructed in such a way that, first, the semi-finished product being machined is not subjected to decentering, and, secondly, that the force required to set the semi-finished product in rotational motion is transmitted more uniformly.

And indeed, we see that the form in which this idea is realized in historical sequence is the direct transition to the installation of a rear movable headstock of the same construction as the front one, i.e. with a rotating faceplate instead of a dead center and receiving this rotation from a common shaft.

Thus, the axial recesses in the left-hand part of the semi-finished product are not worn away, which preserves the correct setting of the axis, and the rotational force is transmitted uniformly to both ends of the semi-finished product.

A wheel lathe of this higher type, schematically represented in comparison with the preceding type, is shown in Fig. 2.

Fig. 2.

Here, on the bed S, two headstocks B are installed, of which one, the front one, receiving motion from the drive shaft or motor, is stationary, while the other, the rear one, can move along the bed S. The headstock usually consists of the headstock body B with plain (old designs) or split bearings E, in which is placed the spindle shaft A, ending in a center d. On the shaft A is fastened the faceplate C with a toothed rim. The semi-finished product P is mounted on the spindle centers d and is set in rotational motion by the drivers D.

By shifting one of the headstocks and one or both spindles with the center d, the lathe can be adapted for semi-finished products with axles of various lengths, i.e. for broad- and narrow-gauge wheels, or for car and locomotive semi-finished products with crankpins.

How expediently this scheme has been realized, representing at the present time the most widespread type of wheel lathe, I shall show below, in a detailed examination of the designs in their historical sequence; for now I shall proceed to two other schemes.

When turning tires, it is important that their machined surfaces be exactly concentric with the axle of the wheelset and thereby concentric with the surfaces of the axle journals. It is not always possible, however, to rely in this respect on the spindle centers; and so the designer has the idea of eliminating a possible error, i.e., of making the rotation of the wheelset independent of the centers. This idea was first realized, as we shall see below, more than 40 years ago. A machine based on this idea is constructed in such a way that the wheelset can be mounted and centered not on the spindle centers, but on the axle journals.

The diagram of such a machine may be taken as Fig. 3. Here, first of all, we see that in the fixed sockets D of the headstock beds there are mounted non-rotating spindles B, capable of performing only a translational motion. The thrust centers d serve only to prevent the axle of the wheelset from moving to the right and to the left.

Fig. 3.

The wheelset is mounted by its journals in the bearings C, which are centered by the bolts a. The faceplates R, with toothed rims, rotate on the bushings A, which are continuations of the headstock beds, and receive motion from the main shaft V through the gears Z.

One of the two headstocks—the front one—is fixed; the rear headstock can move along the bed. When turning, the wheelset, while bearing against the centers d, rotates by its journals in the bearings C. This, however, does not exhaust the scheme. The obvious drawback of such fastening of wheelsets—the abrasion and possible scoring of the axle journal—forced the designers to attach the bearing to the faceplate and make it rotating, i.e., to convert it into a retaining bushing.

Finally, the fourth type of machine, found only in America, is shown schematically in Fig. 4. First of all, the machines

of this type have the aim of eliminating a very substantial shortcoming of the two preceding types, consisting in the fact that the engagements of the gear rims with the pinions on the main shaft of the machine

Fig. 4.

can never be fitted exactly. As a result, first one gear, which enters into engagement earlier than the other, is worked loose, and then the other as well. Machines of the 4th type, incidentally the most powerful ones, receive their rotary motion from a single gear wheel R and transmit it to the axle by means of the dogs of the faceplates S₁, connected with R. The headstocks B, both movable—an essential difference from the preceding arrangements—carry faceplates S with thrust spindles T, ending in self-centering shoes n; the faceplates S likewise secure the axle by means of the dogs q and rotate together with the spindle shaft.

All the varied designs of wheel lathes fit into these four arrangements.

Let us now pass to the second principal criterion for evaluating a machine, namely to the type of working tool used for turning.

The working tool, like the very first design of the wheel lathe, was taken from the ordinary lathe, i.e., it was an ordinary cutter. By adapting the working tool to the character of the surface being machined, a form tool was created for shaving the tire. But alongside the cutter, which removes chips with a single cutting edge, there existed an attempt to introduce also a more complex working tool—the milling cutter.

This attempt, as we shall see below, even assumed definite real forms.

The productivity and cleanliness of the machine’s work are greatly influenced by the mounting and regulation of the movement of the working tool;

therefore the design of the slide-rest, in other words, the setting and movement of the cutter along the line of machining, is the third criterion in the assessment of a wheel lathe. Depending on the feed of the slide-rest cutter, there are two kinds of slide-rests: first, simple slide-rests, on which work is performed with plain and forming cutters; and second, automatic ones,*) by means of which the shaped profile of the tire is machined with a plain cutter only. In the first case the worker must necessarily regulate the thickness of the shaving by hand, whereas in the second both the width and the thickness of the shaving are determined by the automatic travel of the slide-rest, although slide-rests of the second type are designed so that the travel of the cutter can also be regulated by hand.

Finally, the third type of slide-rests consists of slide-rests for the milling cutter.

II.

We shall, of course, characterize the machines themselves in the same order as was indicated in the preceding chapter.

At the same time, when speaking of the method of imparting rotary motion to the wheelset, we shall have to give an account of the designs of the headstocks in all their essential details, beginning with the spindles, spindle shafts, and faceplates, and ending with the method of moving them in the longitudinal direction. It is also important to dwell on the methods of changing and regulating the speeds of rotation. As for the bed of the machine itself, it will be more convenient to speak of it when describing the machines.

It is unnecessary to dwell on machines of the first type, if only because, from the design point of view, they represent merely an ordinary lathe on an enlarged scale. We therefore pass on to the description of the second type and shall try to trace the development of its design.

One of the oldest machines of this type is undoubtedly the machine of the English works Sharp, Stewart & Co. in Manchester, once famous throughout Europe. One of these old-timers has been working to this day, since the time of its construction, on the St. Petersburg premises (wheel shop) of the Petersburg–Warsaw Railway. Evidently, this is one of the first wheel lathes built by the firm Sharp, Stewart & Co., which no longer exists; and its age is not less than 50 years, since it was working even before its installation in the workshops.

I have not been able to establish more precisely the date of manufacture of this machine, since in the literature on Stewart’s first machines there is no informa—

*) Perhaps the earliest design of an automatic slide-rest (turning according to a shaped template) for a lathe is to be found in Prof. A. Ledebour’s book Die Verarbeitung der Metalle auf mechanischem Wege, 1877.

no data are available either in the archives of the Loudon Brothers, Ltd. works, which inherited the defunct firm.

The essential features of this machine, undoubtedly the progenitor of wheel lathes of the 2nd type, are its spindle—

Fig. 5.

—shaft and faceplate. The spindle shaft C, Fig. 5, as in old lathes, is turned to a taper at its journal portions, which enter the blind bearings B. The adjustment of the shaft, when the integral brass bushes V and V₁ wear, is effected, as in lathes, by round nuts m and n (diam. 30 m/m, triangular thread, screw pitch 5 m/m).

The spindle D, ending in the thrust center S, has a guide piston b, made in one piece with D, which also prevents longitudinal bending of the spindle. The flange a keeps the spindle from longitudinal movement by bearing against the nut E, screwed into the spindle shaft from within. The screw c of the spindle, turned by a flywheel, moves the center S forward or backward by means of the center-holder K, which is kept from rotary motion by the knob r with a pin running in the groove q of the center-holder. The liner e of the center-holder K is of steel, as is the spindle; the spindle shaft C is of cast iron. The faceplate P, cast in one piece with the toothed rim (a feature of the old design of the machine), is connected to the spindle shaft by six bolts; it receives its rotary motion from the shaft V, passing beneath the headstocks along the axis of the machine, through the gear Q, which can be disengaged by the fork R, if required. In Fig. 6 a photograph of the movable headstock is shown, and here the blind bearings are clearly visible. The front fixed headstock carries the transmission for rotating the shaft and a separate stepped pulley, with a gear on the same axis, meshing with the toothed rim of the faceplate. The latter is provided in order to obtain the high speed of rotation of the machine required when turning axle necks.

Fig. 6.

The shortcomings of this design are obvious: blind bearings, and hence the difficulty of assembling and, chiefly, adjusting the spindle shaft when the liners have worn, a difficulty further complicated by the one-piece liners; the second essential drawback is the one-piece casting of the faceplate and toothed rim, since in this case the breakage of several teeth

forces the entire faceplate to be discarded. Soon, however, already in the 1860s, Stewart’s works designed split bearings, while the toothed rim was cast separately from the faceplate and fastened to it with bolts. In this case, the breakage of several teeth makes it possible either to fit the chuck freely onto the rim removed from the faceplate, or, in the extreme case, to cast a new rim if its repair is impossible.

The highest form of the design is the lathe of Engineer Mesmer (machine-building works in Strasbourg), built in the 1850s. Here, evidently, Mesmer quite consciously eliminates the shortcomings of the lathes of the 1st scheme and Stewart’s 2nd. Leaving aside for the present a complete description of the lathe, let us note only that here the bearings are split; the liners consist of two halves, and the toothed rim is cast separately and fastened to the faceplate with bolts. The right headstock (Plate 1) carries (fig. 22) a sliding spindle C₁, whereas the left has a fixed C (fig. 20) with center e. The faceplates J and J₁ are set in rotary motion from shaft K through gears k mounted on it (fig. 17), while shaft K receives its motion through gear train G, L and K₁ from the stepped pulley D (figs. 17 and 18). From fig. 18 on the left headstock it is evident that, by engaging or disengaging gear d, two different speeds can be obtained: a low one—for turning the tire—and a high one—for turning and polishing the axle journals.

With certain design naïvetés (the dead center f of the fixed headstock with the possibility of small displacements of the spindle shaft, secured in a boss of the bed—fig. 17) this lathe nevertheless is distinguished by great thoughtfulness, and according to its scheme lathes of the 2nd type were designed for at least 40 years from the time of its invention.

If we compare Stewart’s lathe with Mesmer’s, then, in terms of constructive improvements, it is easy to establish the continuity of Stewart’s ideas, although I have not been able to establish this precisely in chronological order.

Thoughtful detailing of the design according to one idea or another is a characteristic feature of German engineers, and all the details of Mesmer’s lathe are expressed so vividly that for a long time they formed the basis for the design of lathes of this type.

A direct development of the design of Mesmer’s lathe is the very widespread lathe (figs. 7 and 8), manufactured even now not only in Russia but also in the West.

In figs. 7 and 8 are shown the left and right (movable) headstocks of a lathe of the B. K. A. Weichelt works in Moscow. Comparing this lathe with Mesmer’s, we see that in it, first, the sliding—

Wheel Lathe

Fig. 6.

Fig. 7.

new spindle, as well as at the fixed headstock; secondly, there is a device for the automatic feed of the carriage (gears e with cranks f are visible on the right-hand headstock, and h on the left); and, thirdly, the movable headstock is fed not by hand and levers, as in Stuart and Mesmer, but by the screw K, entering a screw self-acting nut connected with the sole of the headstock. This machine of German origin is varied by European and American works only in the sense of improving individual details.

As one of the substantial shortcomings of this machine it may be pointed out that the stepped pulley is mounted on the spindle shaft. In order to avoid overloading the spindle shaft, which causes wear of the bearings, it was necessary to avoid placing the stepped pulley on the spindle shaft. And, indeed, the constant wear of the bearings of the fixed headstock compelled designers to move the stepped pulley out and to secure it in separate bearings, which is what the Weichelt works now does. This was the essential modification of the machine. Further, the fastening and centering of the spindle center-holder are also unconstructive, as in Mesmer’s machine. In working out the design of the spindle, the factories gave several more or less successful solutions to the problem (Schiess in Germany, Felzer in Riga); but the spindle of the Br. Bromley works in Moscow is distinguished by excellent constructive thoughtfulness.

In Figs. 9 and 10 are shown the ends e of the spindle and of the spindle shaft, which lies in the bearings of the headstock frame. The center-holder, which in Weichelt’s design reams out the hole at the end of the spindle shaft—V, here (Fig. 9) is centered by a conical bushing B (bronze), which is tightened by the nut L, screwed into the washer G, fastened to the tail of the shaft. It is clear that, instead of the worn nut B, a new one can easily be installed; but if the end of the spindle shaft wears out, the whole shaft has to be thrown away.

Fig. 9.

In Fig. 10 is shown the thrust end of the shaft. Here K is a steel washer, L are round nuts for tightening and fastening the spindle shaft (the principle of an ordinary lathe), D is a screw-in bronze liner for the screw B, feeding the spindle; I—

bushing, and R—a washer, both of steel, for holding the spindle in a definite position, and finally S—a thrust center for taking up the axial pressures occurring when turning the axle-blank.

Fig. 10.

This, in essence, may conclude the characterization of the 2nd type of machines in their modern form. It remains only to note that the Americans do not make separate thrust centers like S, but transmit the axial pressures to the bearings of the spindle shaft, reinforcing the tail thrust of the spindle screw. However, it can hardly be acknowledged that there is any expediency in this design, since, with the enormous power of their machines—up to 50 HP—this must cause wear of the bearings. Although, on the other hand, the method of securing the axle-blank in which the axial pressure is transferred to the entire spindle shaft, as we shall see below, considerably weakens the shortcomings of the thrust in the American design.

The attempt to construct machines according to the 3rd scheme also belongs to very remote times. As early as the Paris World Exhibition of 1867, machines with “bears” were exhibited; and a circumstantial description of such a machine from R. Hartmann’s works in Chemnitz, exhibited in 1873 at the World Exhibition in Vienna, is found in E. Hesse’s report on the exhibition. Very soon, however, the method of securing the axle-blank in its bearings by its axial necks, with the immobility of the thrust centers, proved unsuitable, and German works long ago ceased making such machines. And only among us, 20 years after the Germans had discarded this design,

factories make these machines according to the old German types, continuing to offer them even now.

In exactly the same form (apart from the supports) as it was exhibited at the Vienna Exhibition of 1873, the machine is shown in (Plate 2), as built by the Bromley Brothers’ works.

Here the bearing P is fastened to a bushing that serves as a continuation of the headstock bed. The bolts B, below and above the bearing, make it possible to press and center the axle journals. From the stepped pulley, the gear wheel Z receives rotation through the gear Z₁, transmitting motion to the shaft geared with the teeth of the faceplates. By the screw K, the axis of the stepped pulley sets in motion the screw wheel v with a crank for the automatic feed of the supports. How impractical the fastening of the wheelset in the bearings is may be judged from the fact that in all workshops where these machines are used, the bearings have been removed from them, and the wheelset is set up on non-rotating centers. This is understandable, since it is impossible to center the axle of a wheelset by such a primitive method. In the best case this arrangement did not give a correct turning of the tires, and in the worst case the axle journals were scored.

Thus, the realization of the idea of setting the wheelset by the axle journals in bearings must be recognized as extremely unsuccessful. The Americans solved this problem much more successfully, replacing the bearings with dogs that grip the journal and rotate together with the wheelset.

A machine of this type was first exhibited at the World’s Fair in Chicago in 1893, and in Russian literature was described by Prof. Gatsuk in his book on American machine tools for metalworking). Since that time the construction of these machines in American manufacture has remained almost unchanged, and Figs. 1 and 2 (Plate 3)* represent an exact copy of this machine, produced by the Moscow works of B. K. A. Weichelt.

The bushing W, being a continuation of the headstock bed, at the same time serves as the axis for the chuck Z rotating on it, which receives motion from shaft L through gear Z₁. The chuck is held on the flange of the bed by the angular washer K. The spindle V, ending in the center holder T, serves as the axis of the movable cone fitted on the cone of the faceplate R with four dogs Q, which grip the journal of the wheelset. By rotating the roller by the handwheel fitted on E, motion is transmitted to screw A, which receives this motion from the toothed flange B through gear C. When A rotates, the faceplate R moves forward and lowers onto the journal the dogs Q, which are restrained from motion along the axis by the cup m. The washer n, placed on screws, [[unclear: continues on next page]]

) A. Gatsuk. American Machine Tools for Metalworking*. St. Petersburg, 1896.

covers this entire system of the sliding carriage, leaving an opening for inserting the axle journal. If the washer n is removed, the entire carriage system is easily taken out, since it is freely mounted on the center support T; in this case the lathe is converted into a neck lathe, although in fact it is not used as a neck lathe. In Fig. 2 (Plate 3) a section of the bed is shown. S—foundation plate; S₁—headstock bed; O₁—axis of the stepped pulley; O—shaft axis; O₂—axis of the chuck; K—cock for draining used lubricant.

In this lathe it should be noted that the shaft O is carried outward, whereas usually, in designs of the second type, the Americans and the great majority of European works place these shafts under the headstocks, thereby making access to the inspection of the gear transmissions difficult.

The feed of the spindle along its axis is effected in the usual way: the handwheel M is turned, and the screw roller V advances the center support T, which is kept from rotary motion by the key F. On the end of the main shaft a wheel is mounted with a shifting crank pin for transmitting, through the roller O (Fig. 11), the automatic movement of the supports by the usual ratchet coupling R. In our country work is usually done on centers, whereas the Americans put caps a on the centers, to which the axial pressure is transmitted, as to thrust pads.

Fig. 11.

As I have already said earlier, the Americans had been making this lathe since the time of the Chicago Exhibition without substantial design changes.

…of the former, having discarded only the driver and replacing them with special cams, which they apply to all types of machines.

But this will be discussed below; for now let us proceed to the description of a machine of the same type, manufactured in Russia by the Riga works of G. Felzer according to a somewhat modified—one may say improved—model of the German works of Frust Schiess in Düsseldorf.

Like most German designs, this machine is distinguished by excellent detailing, although in essence it introduces almost nothing new into the American design. In Fig. 12 a fixed headstock carries the chuck P, ending in the cup a, intended for the axle journal of the wheelset. The chuck rotates on the bushing N, fastened to the bed with countersunk screws, which is far more constructive than in the American machine, where the bushing is cast integrally with the bed; moreover, the chuck is additionally supported by the bearing Q; the flanged part of the chuck with the cams holding the wheelset is likewise fastened with studs. Inside the hollow bed, where there is convenient access through a hatch, a non-rotating spindle is installed, whose feed motion is obtained very conveniently: the axis of the bevel gear M projects outward, ending in a small handwheel at the chuck, under the hands of the turner. By rotating the handwheel, this motion is transmitted to the shaft V by the bevel gears M and L; further, by the gears B and B₁, connected by a Galle chain, they impart …

Fig. 12.

…stepwise motion to the spindle. The spindle bearings, as well as its support C, are placed in the castings of the headstock bed.

Fig. 13.

The rear movable headstock (Fig. 13) is moved along the bed of the lathe by hand, by means of a lever with a ratchet, fitted onto B. B terminates in a gear that drives the headstock along a rack fastened to the bed in a vertical plane. This lathe, of 15 HP, is driven by an electric motor, which transmits motion to the toothed change-gear by a Renold chain, and can give six different speeds of rotation. The half-tire is fastened by means (Fig. 13) of 8 screws, which fit the bushings onto the axle journals, and 8 that press the lug against the inner sides of the wheel rims. A carriage for receiving and setting the half-tire is also placed on the bed of the lathe; but since this has no direct relation to the construction of the lathe, I shall not dwell on it. As in the American lathe, the main shaft for transmitting rotation to the chuck is brought out from under the headstock, which is very convenient as regards access to it.

With this one may conclude the illustration of the scheme of the 3rd type of lathes, of whose comparative merits and shortcomings I shall speak below.

If the American lathes of the second category of the IIIrd scheme trace their genealogy to European inventions, then the lathes of type IV, according to the scheme, are already purely American.

Seeking to increase the productivity of lathes, American engineers pay serious attention to securing the wheelset. With their system of machining—taking a heavy chip during the roughing pass and a wide finishing cut—this is essential. Since the general considerations have already been set forth at the beginning of the monograph, we shall proceed directly to the description of this construction, the only one of its kind, manufactured by The Pond Machine Tool Co. in Plainfield, New Jersey.

This lathe, also first exhibited at the Chicago exhibition, was described by Prof. Gatschuk, and since 1893 has changed little.

Fig. 14.

As can be seen from Figs. 14 and 15 (Fig. 15 shows the installation in the shops of the New York Central Railroad), the wheelset is placed in the middle fixed headstock, which constitutes a bearing in which a sleeve rotates, ending in faceplates a with four dogs (Fig. 14). Mounted on this sleeve with the faceplates is an enormous wheel P with spiral teeth.

Both end movable headstocks T, with faceplates P instead of spindle centers, are provided with conical bushings that slide onto the similarly conical shoes A—B, which tightly embrace the axle journals (Figs. 14 and 15). Thus, centering is effected by advancing the end headstocks. When the end headstocks have been set, the wheelset is secured by the dogs of the faceplates P and P₁, which grip the wheel rim from the inside. The wheelset receives rotation from the faceplates a—P, which are set in motion by the central gear wheel P (Fig. 14). To allow the wheelset to pass through, a cutout is made in the bearing, in the sleeve, and in wheel P, with a removable piece A, Fig. 15, inserted in place during operation. The entire lathe (the carriages are located on the rear side) is set up so that the height of its cent—

the level is the same as that of the axle of the wheelset; therefore the wheelset is simply rolled into the machine along a rail track, and for this machine no special lifting mechanisms are needed. Of course, on this machine only wagon wheelsets are turned.

Fig. 15.

It remains still to dwell on the fastening of the wheelsets, practiced by the Americans, since the traditional drivers and dogs provide too weak a fastening for powerful American machines.

In Fig. 16 is shown the chuck of an American machine of the 3rd type, with dogs of special construction. These dogs run in corresponding grooves of the faceplate and consist of three parts: the body of the dog K and two wedges T and L with perpendicularly arranged axes. The wedge, or jaw T with a notch, can come out of its socket if the wedge L is struck. When the wheelset has been set in place, the dogs are moved out so that the jaws T are opposite the side surface of the tire; then, by striking the wedge L with a small hammer (a hand hammer), T is moved out until it touches and even presses upon the tire of the wheelset; after this, with bolts whose heads enter the grooves of the faceplate, the wheel is tightened, pressing it against the jaws T. This system is good in the respect that it really firmly holds the wheelset—

to release; but both these cams and the cams*) that grip the wheel rim from the inner side have the very substantial drawback that it is impossible to give them equal pressure. Consequently,

Fig. 10.

with this method of fastening there will always be skewing forces, and the pressing of the spindle shaft or of the chuck bushing will be produced on one side, which causes uneven wear of the shaft and of the bearings. This drawback is not present in the old method of fastening the half-skid by means of drivers; but the old method is not suitable for powerful machines that produce a heavy and wide chip, because the spindles alone cannot withstand such a load as 30–45000 kg on one cutter.

Thus, the four schemes of machines described above fully exhaust all existing designs.

*) These cams, i.e., the second ones, are installed on machines of the IV-th scheme. In general, this method of fastening is practiced by the Americans also for machines of the II-nd scheme.

III.

In passing to the appraisal of the working tool, its setting and the regulation of its motion according to the turning profile, it is necessary first to review and assess all types of beds of wheel lathes, because the carriages, being mounted on the bed in the span between the headstocks, require a structurally convenient bed.

In Figs. 1 to 4 (plate, Fig. 17) all existing plans of beds are shown, and in Figs. 5–6 their transverse sections, which I divide into two groups, taking as the basis of classification the position of the main shaft in relation to the headstocks. Fig. 1 gives an idea of the beds of old machines; here the widening of the carriage plates $P$ is not very great, which, incidentally, was not required in weak old machines. Fig. 2 is the American type of bed for machines for permanent rolling stock; the bed has projections for the carriages. Fig. 3 is the ordinary modern type of bed, and its part $GDEF$ is often cast separately and bolted to the main plate $ABCH$. Finally, Fig. 4 is the bed manufactured by the Weichelt works, with the attached part $BCDE$.

However reliable the bolting together of two parts of the bed may be, such a bed must nevertheless be considered unsatisfactory, because the lowering of $GDEF$ (Fig. 3), or of the greater part of it, is quite possible. The use of the bed of Fig. 4 is completely incomprehensible, where the part $CDEF$ is undoubtedly superfluous. In Figs. 3 and 4, $P$ denotes the cast-iron plate common to the two carriages, moved with them along the bed; $S$ denotes the carriage for turning the tread profile, and $S_1$ the carriage for turning the side surfaces of the tires. It is clear that, in the sense of longitudinal movement of the carriages, the arrangement of Fig. 4 is also unsatisfactory, since here each carriage has to be moved separately, whereas in the arrangement of Fig. 3 both carriages are moved simultaneously for each wheel.

Fig. 17.

In Fig. 5 we see that the shaft is placed exactly under the headstocks, which is extremely inconvenient for observing the meshing of the gears, for cleaning and repairing the machine; therefore the arrangement according to Fig. 6 is more preferable, since the shaft is brought out from under the headstocks and is quite accessible for inspection and repair.

Thus, the bed according to the type of Fig. 3, combined with the type of Fig. 6, must be recognized as the most constructive, if it is cast ...

sufficiently strong. In such a case, setting up and moving along the bed are most convenient for the carriage, because it is not threatened by settling or skewing from the settling or skewing of the coupled parts of the bed.

Speaking of the working tool and its mounting, it is best to begin with the milling cutter, as standing apart from the other types of tools.

In Russian literature, milling wheel lathes were first described by Prof. Knabbe *). As we shall see below, in their construction these machines do not differ in any way from the second type of wheel-turning lathes, although with equal success they could have been constructed according to the 3rd and 4th types. Their chief difference from the turning type of wheel lathes is the very low speed of rotation: \(15^{m}/_{m}—45^{m}/_{m}\) per minute, whereas in wheel-turning lathes it reaches up to \(4500^{m}/_{m}\). But the milling cutter, or rather the cutter edge, machines the half-tire in one revolution.

Fig. 18.

The milling carriages, in the form in which they were designed by Roth and built by the German Schiess works, are shown

*) Prof. Knabbe.—Modern equipment of machine-building works and railway workshops. Kharkov, 1896.

in Fig. 18. Each carriage is equipped with three milling cutters: one form cutter, shaped to the profile of the tire, rotating about the axis V (the left carriage), intended for turning the running surface and the flange of the tire; the other two, with axes P—P (the right carriage), are intended for turning the side surfaces of the tire. The system of these cutters (on the right carriage the form cutter is shown, while on the left the side cutters have been removed in order to show the transmission of motion more clearly) receives motion from the pulleys NN and OO, which rotate the axes P—P by means of the gears S and T and the screw engagements Q and R, while the form cutter—the axis V—is driven by means of the bevel gears W—X and the screw engagement Y—Z.

Although, in comparison with the productivity of wheel lathes of that time, milling lathes did give greater productivity, according to the testimony of the Schiess works, there were no scientific tests of these machines.

In Russia the only works (the Bromley Brothers in Moscow) built only one milling lathe for the Ussuri Railway, and the matter went no further. In Europe they are now no longer manufactured at all; and America, which makes especially extensive use of milling, has taken an entirely negative attitude toward the milling cutter for the wheel lathe. This is understandable, because the milling cutter is too delicate and too expensive a tool for such rough work as turning tires.

Thus, the milling turning of tires on wheelsets is at the present time a possession of history and represents only an illustration of the fascination with the milling cutter.

Passing to the description of carriages with cutting tools that work by the ordinary removal of chips, it is necessary first to dwell on the material of the cutters.

As early as 1865, in England and America (the United States), the so-called self-hardening steel began to spread, patented in England by Robert Mushet. The chief property of this steel consists in the fact that it does not lose its hardness when self-heated even up to 300° during the cutting of chips. However, the increase in the speed of operation and the strengthening of machines working by cutting chips required greater hardness and strength of the tool. And so, in 1900, at the World’s Exhibition in Paris, the American steel-casting works Bethlehem Steel Works exhibited steel cutters made of unusually hard self-hardening steel, which during operation heated up to a dark-red color and did not lose their hardness.

This steel was obtained by the American engineers White and Taylor. Self-hardening steel in general is characterized by a greater content of tungsten, and if one compares steel

If one compares Mushet’s steel with Taylor’s, the chemical composition will be expressed as follows:

	Mushet	Taylor & White
C	2%	1%
Mn	2.5%	0.17%
Si	1.3%	0.2%
W	5%	9%
Cr	0.50%	3%

Thus we see that Taylor and White’s steel contains almost twice as much tungsten (5% and 9%).

However, the secret of the improved quality of this steel lay chiefly in its heat treatment, i.e., in the method of hardening. The method of obtaining self-hardening steel is now no longer a secret, and it is being manufactured with considerable success even at our largest works (Putilov).

The invention of high-speed cutting (self-hardening) steel produced a complete revolution in the field of machine building, forcing designers to strengthen machines considerably and to increase cutting speed. How excellently this steel works may be judged from the tests by J. M. Gledhill of steel of grade A. W (Armstrong Whitworth works), who says that, with a feed of \(0.4 \text{ m/m} \times 20 \text{ m/m}\) (thickness and width of chip) and a rotational speed of 46 meters per minute, the cutter worked 7–8 hours without regrinding; in turning oil-hardened gun steel (strength 70 kg per 1 square millimeter), with a feed of \(2.3 \text{ m/m} \times 4.5 \text{ m/m}\)—17 m/m—the cutter worked 9 hours without regrinding. The investigations by engineer Georgov (St. Petersburg workshops of the St. Petersburg–Warsaw Railway) on “Novo” and “Diamant” steel in the work of wheel-turning lathes likewise gave excellent results*), and the wheel shop knows of no other cutters than those made of self-hardening steel, avoiding the use of this steel only on weak wheel lathes that cannot withstand a heavy chip. The cutters, taking the heaviest chip that the machine could withstand, worked for 17 days, daily from 3 to 8 hours, and gave the following results:

The “Novo” steel cutter before work weighed \(8 \tfrac{7}{8}\) lb.; length 355 m/m
” ” ” after ” ” \(8 \tfrac{5}{8}\) ” ” 347 m/m
The “Diamant” steel cutter ” ” \(4 \tfrac{1}{8}\) ” ” 334 m/m
” ” ” after ” ” 4 lb. ” 320 m/m

Thus, during about 100 hours of work, the “Novo” steel cutter lost only \(1/4\) lb. from sharpening the cutter, and “Diamant” \(1/8\), while a considerable part of the cutter material went into sharpening it;

*) The investigations were carried out in 1901, but have not been published. They are in my possession in the form of manuscript tables kindly supplied to me by G. A. Georgov.

and not as a consequence of wear of the tool. In general, much testing of high-speed steel has been carried out in recent times, and among these experiments stand out the extensive investigations of J. T. Nicolson, professor of the Manchester Municipal School of Technology), who worked on determining the most advantageous speed of rotation, feed, and cutting angle. No less interesting are the tests of the American engineer W. Taylor, who used speeds of rotation from 80 to 150 meters per minute with chips from \(12.5^{m}/m \times 1.5^{m}/m\) to \(2^{m}/m \times 0.7^{m}/m\); moreover, the work was carried on for 1 h. 30 min. without resharpening the tool*).

All these investigations show that there is no point in using any material other than high-speed steel for the tools of a wheel lathe.

As to the form of the tool, it is of two kinds: 1) a plain tool, the ordinary lathe tool, and 2) a form tool according to the tire flange or to part of the rolling profile. A plain tool is usually mounted on automatic carriages, and form tools on simple ones.

We shall now turn to the characteristics of the carriages.

The so-called cross carriage, giving the tool two mutually perpendicular directions of motion, was altered chiefly in the direction of strengthening the construction. In Figs. 24 and 25 (Plate 1) there is shown a German carriage of weak, slow-running machines of the end of the 1840s. Here \(p\) and \(q\) are screws for moving the carriages \(S_1\) and \(S\), which carry the tool: the first perpendicular to the axis of the wheelset, the second parallel to it. The carriage \(S\) turns about its axis, being mounted on pin \(n\). When it is necessary to set screw \(p\) parallel to the rolling profile of the tire, the bolts of carriage \(S\) are loosened and it is turned on pin \(n\), after which the bolts are again tightened. By the bolts of carriage \(S_1\) the tool \(r\) is held: the bed of the carriage is fastened to board \(P\), drawn to the machine bed by bolts \(m\), which are loosened when it is necessary to move the carriage board along the bed. Board \(P\) is moved along the bed by gear \(t\), meshing with toothed rack \(u\), fastened to the bed of the machine.

The defects of this carriage are obvious: first, the bolts holding the tool require a thread in special cut-outs made for them in carriage \(S_1\); this is very inconvenient, since when the thread in the notches is worn out it will be necessary to increase the diameter of the new thread, or even to discard carriage \(S_1\) altogether.

The second defect is pin \(n\); although this connection is simple, it is too weak and unreliable for more powerful machines.

) See Engineering*, 1903.

) American Machinist, 1907, March 2–9. — “The Shape and duty of roughing tools,” W. Taylor.

Stewart’s support is more constructive. Instead of the high carriage $S_1$, here the carriage A (Fig. 19) is made with slots for the bolt heads; the cutter is secured by crosspieces tightened by the nuts of the bolts. Consequently, if the thread of a bolt is stripped by overloading the cutter, it is necessary only to install a new bolt. Both carriage A and B can rotate about their own axes, and the carriages are mounted not on pins, but on broad heels cast in one piece with the carriages and entering corresponding recesses: for A in carriage B, and for B in the frame of support S; the flanges a and b of the carriages are tightened by bolts, as in the German design of the support. The whole support, together with the frame, is moved perpendicular to the axis of the wheelset along plate D by means of a gear mounted on axis C and rack R.

Fig. 19.

Although this support is of very ancient origin, nevertheless all its details are so well considered that the modern construction of the cross support differs in almost no respect from it, as we see, for example, in the designs of American and English supports.

Before the designer’s thought turned to automatic supports, attempts were made to increase the productivity of machines by installing several cutters at once, taking in the entire profile of the turning. Such, for example, is Erhardt’s support (Fig. 20).

But these supports required powerful machines, and moreover the breakage or dulling of a single cutter stopped the work, thereby slowing the turning. Soon these supports were completely abandoned.

Fig. 20.

Much more complex and more varied are the supports of automatic action. The essence of the construction of such a sup-

The Wheel Lathe

...of the support consists in giving the cutter an automatic movement along the profile of the tire. Let us try to examine the construction of these supports, adhering to the historical sequence of their appearance.

In Figs. 1, 2, 3, and 4 (Plate IV) we have an automatic support of German origin. This support, first introduced in Russia by the Gerlach and Pulst works, is also built by Weichelt and by several large works and some smaller works.

The principle of the automatic action of this support, as of other automatic supports, consists in causing the cutter to describe the line of the tire profile, composing this curve from two motions: the movement of the lower carriage in the usual manner and the automatic movement of the carriage carrying the cutter according to a definite template. The support consists, properly speaking, of two separate supports carrying cutters—one for the flange, the other for the running surface of the tire; the lower carriages stand on common parallel ways. By the screw E (Fig. 3), with a right-hand thread for carriage P and a left-hand thread for O, these carriages are moved parallel to the axis of the half-slide; by gears K the worm J (Fig. 1) is set in rotation, transmitting rotation to the worm wheel L, on the axis of which the template H is eccentrically mounted; into this template enters the finger of the sleeve of the upper left-hand carriage P. Another template G is fixed immovably on the support frame for carriage O, which feeds the cutter for machining the running surface of the tire. Since the slope of the running profile is small, during the longitudinal movement of the lower carriage the carriage O, carrying cutter 2 (Fig. 3), is drawn along by the finger entering the template, notwithstanding the immobility of the template, and thus describes the turning line. In addition to the automatic feed of the cutters, the construction of the supports, as may be seen from the drawings, makes it possible to move the cutters by hand with screws g and p (Fig. 3). The template for the flange is a closed curved groove H (Fig. 3); for the running profile it is of the kind shown in Fig. 4.

Although, in comparison with others, this support works satisfactorily, nevertheless it cannot be recommended, because the template in the form of a closed curve wears out rather quickly, and it is rather difficult or impossible to make it by one’s own means if the workshop has no universal milling machine.

Without essential changes in the construction of the body of the preceding support, but with the closed-curve template replaced by a template of a more suitable type, there is the support in Fig. 5 (Plate IV), which is very widely used. These supports are usually placed on the foundation support plate P, and are moved by means of screws V. The carriages move parallel...

the half–carriage by the screw t with threads in opposite directions. The shifting of the cutters a and b is effected by means of the template A₁ for the tread surface n, and of the template A for the flange. In addition, the cutters can also be shifted by hand, by means of handles put onto the squares m₂ and m₁.

The advantages of this tool carriage consist in the convenience of replacing a worn template for the flange, which is easily made by one’s own means.

Before passing on to the subsequent designs of automatic tool carriages, I shall point out a very substantial shortcoming of the two preceding ones.

For the proper turning of the flange, not only the correctly described curve of the profile is important, but also the uniform feed of the cutter.

If, however, we trace the feed of the cutter in these tool carriages, we shall observe that there is no uniformity of feed here. In fact, the resulting displacement of the cutter at the beginning and at the end of the turning of the flange will be large, while at its apex it will be considerably smaller, because the displacement of the cutter caused by the template, i.e. in a direction perpendicular to the axis of the half–carriage, is the projection of the template curve onto the axis of displacement.

But with one and the same angular displacement per unit time (Fig. 21), the displacements of the cutter a₁ b₁ and AB will be different, since the radii of curvature of the arcs ab and AB are different.

The template curves of the first and second designs of automatic tool carriages have different radii of curvature along their length; consequently, nonuniformity of the cutter feed is inherent in them by the very nature of the design. Graphically this feed may be represented as in Fig. 22.

Fig. 21.

To eliminate this shortcoming, A. Schubert attempts, in designing an original template (Plate V, Figs. 1–6). The screw 1 (Figs. 4–5) sets in rotary motion the screw bronze gear 2 with the axle 3, passing through the phosphor-bronze bushing of the carriage 6 (Fig. 4) and ending in the heel 4 in the stationary template 6 (Fig. 6); on axle 3 is mounted the gear 5 (Figs. 4 and 6), engaged with the pins of the template, which is made of the hardest cast iron. The screw

Fig. 22.

rotates by means of an ordinary ratchet transmission (Fig. 1). Moving over the template, gear 5 draws carriage 7 along with it and at the same time causes carriage 8 to move along an axis perpendicular to the axis of the half-slide (Fig. 4). The screw transmission and the engagement of the gear with the pins of the template are in an oil bath.

If turning according to the template is not required, then coupling 12 (Fig. 2), which connects shaft 1 by means of a key with the coupling of gear 14, is disengaged, while gears 14 and 13 are engaged; then the uppermost carriage carrying cutter 9 is fed by screw 10. This is usually done for turning the side surfaces of the tire. If, however, the machine has a slide rest specially for turning the side surfaces, then screw 10 is not installed at all. By screw 11 the whole system is fed by hand. Each of the movements described can be performed by hand, as in other automatic slide rests.

But even by this method the non-uniform feed of the cutter is not eliminated (Fig. 23), although the error is somewhat reduced owing to the stable position of the cutter, maintained at every moment by the rack engagement with the template (Fig. 6, Plate V).

Fig. 23.

The reason remains the same, because the engagement of the rack with the pins of the template gives the exact profile of the tire, but the projections of the displacements of the rack onto the direction of the cutter are different, since the principle by which the curve is obtained has remained the same.

A much more successful design is offered by the firm of Petschler and Glöckner (Petschler und Glöckner, Chemnitz). Both cutters (Plate V; Figs. 7, 8, 9 and 10) a and b receive motion parallel to the axis of the half-slide by means of (Figs. 7 and 8) the ratchet engagement d and screw f. Cutter b, of circular form for turning the flange, receives motion from the engagement of screw g and screw segment h, through which the finger of the link mechanism i passes (Figs. 9 and 10). Lever i rotates about pin s. As segment h is turned by screw g, the link lever i also turns, as a result of which the lower carriage l moves parallel to the axis of the half-slide. But since i, connected with template k, also advances along the radius, carriage m can move in the transverse direction.

Thus, at each given moment the resultant displacements of the cutter can be made equal by selecting the corresponding dimensions of the parts of the mechanism. The feed of the cutter for this slide rest is characterized by the diagram shown in (Fig. 24).

The ratchet gearing of this carriage has cams, by means of which, in addition to the usual method—a lever with a counterweight—it is also possible to vary the amount of feed of the cutter, that is, the width of the chip, by changing the radius of the cam of screw g.

This carriage was patented only in 1904, and in Russia it is not yet being used.

Fig. 24.

There is no doubt, however, that this carriage is the best in construction and will displace automatic carriages that feed the cutter unevenly.

Automatic carriages are widespread in Europe, except for England, and in particular among us (especially recently); in America, however, they are not used at all. The latter is explained by the nature of the turning process, which is so rough in its first stage that the comparatively delicate mechanisms of automatic carriages do not withstand it and quickly fail. The productivity obtained by the Americans on simple cross carriages is so high that acquaintance with their method of turning deserves particular attention.

The process of turning wheel tires and wheel centers among the Americans is the same as in the vast majority of their lathe work. A characteristic feature of this process is that the turning is carried out in two passes: first a roughing cut is made, by which all the unnecessary mass of metal is removed; then—and this is the main point—they turn with a wide cutter to the required dimension or profile. The use of a wide cutter has long been practiced by the Americans. As early as the World Exhibition in Vienna in 1873 they exhibited wide cutters, pointing out the expediency of their turning process, although technically conservative Europe has still not adopted this method of work.

As in the overwhelming majority of cases in which technical questions are resolved by experimentally investigating the problem, American engineers gave a perfectly correct answer to the demands for economizing labor expenditure and reducing time. Indeed, from the latest laboratory investigations*) on the measurement of the work and the forces required for separating metallic chips, we are convinced that the expenditure of work for separating a unit weight of chips depends—

*) Prof. K. A. Zvorykin.

to their thickness, decreasing as the latter increases; moreover, for soft and ductile metals the decrease proceeds more rapidly than for brittle ones. As for broad cutters, the purpose of their use consists in the fact that a wide cutting edge wears down much more slowly than a narrow one, since in the former case less work falls upon a unit of the cutter’s cutting edge for one and the same machined surface.

In view of the high practical value presented by the American method of turning tires, I shall dwell on this method in detail, illustrating it with data from tests carried out in the railroad shops of the New York Central Railroad Company on December 19, 1905*).

The wheel lathe tested had two simple supports, for turning locomotive wheelsets, the largest diameter of which was 2200 m/m. The lathe was of the ordinary, 2nd type (Plate IV), i.e., the type adopted among us and in Western Europe, with the sole difference that the gearing was internal to the faceplate (internal toothed ring). Motion from the motor is transmitted by a Morse chain, and the movable headstock has a separate 5 HP motor for its displacement. The weight of the lathe is about 3150 poods.

The wheelset is set up and aligned on dead centers, and is secured, after additional alignment, by dogs and tie bolts. The cutters, both for the roughing pass and for the finishing turning, are of the best self-hardening steel.

The complete process of setting up and turning is as follows: the wheelset is placed on the centers, putting the crank pins into the corresponding sockets of the faceplate (in our practice the wheelset is usually supported by lunettes, since most faceplates have no sockets for the crank pins). Then the dogs are set so that their knurled jaws fit tightly against the side surface of the tire; for this purpose the jaws are advanced by wedges, which are tapped lightly with a 6–7 pound hammer (a large hand hammer). If, upon checking, a skew of the wheelset on the centers is observed, which is detected by trying with a cutter along the inner side surface of the tire, then by means of the same wedges the dog jaws are adjusted until the wheelset is set correctly on those supports, and then it is tightened with bolts. Next the first chip is taken and the turning diameter is determined. The peripheral speed depends on the quality of the cutter steel and the hardness of the tire and does not exceed 5.5 meters per minute.

*) “Engineer,” 1906, Dec. 28.

The whole process of rough turning is carried out (Plate VII, Fig. 1) in 16 revolutions with the cutter of Fig. 25 in the following order: in 11 revolutions the chip is removed from the rolling surface in one pass.

Fig. 25.

After this the cutters are moved back by hand to the height of the wheel flange and, having set them, with three revolutions they remove the excess protrusion of the flange. Then, feeding the cutter for the 15th and 16th revolutions, they cut off with coarse chips the remaining angular edges of the flange. In this way the process of rough turning is completed without resetting the cutter, and the wheel has a quite peculiar appearance, sharply different from that after our turning.

Fig. 26 (photograph of the cutter, Fig. 25-n.).

Plate VIII shows the wheel after 11 revolutions of rough turning, while Plate IX gives an idea of the size of the coarse chip (scale in inches) and of the form of the broad cutter.

When the coarse chip has been removed, then with 4 profile cutters with a wide cutting surface the turning is finished in 8 revolutions, by

Fig. 27.

Right. Left.

Cutters for finishing during 2 of the first revolutions.

Fig. 28.

Cutter for finishing the tread surface during 2 of the first revolutions.

Left. Fig. 29. Right.

Left. Fig. 30. Right.

Figs. 29 and 30.—Cutters for finishing the flange.

schemes of Fig. 2 (Plate VII), the order of turning being as follows: every two revolutions (see Fig. 2, Plate VII, from left to right) fall to the cutters of Fig. 27, two revolutions to the cutter of Fig. 28, and finally the last pass—2 revolutions each—to the cutters of Figs. 29 and 30. The steady-rest, the bolts by which the wheel is drawn to the faceplate dogs, are loosened, as are the wedges of the dogs, and the steady-rest is removed by crane.

Tests of this lathe were carried out in the shops over a period of 9 hours 6 minutes, and the results of the tests are given in the following table:

Wheel diam.	m/m rotational speed	m/m chip width	m/m chip thickness	Turning time, minutes	Setting-up time, minutes
1664	3978	11,5	11,30	45	8
1664	3848	11,5	11,30	46	9
1664	3900	11,5	12,75	44	10
1664	3640	11,5	13,00	43	9
1664	3822	11,5	12,75	42	8
1664	3432	11,5	13,00	50	8
1664	1632	11,5	11,30	52	8
1664	3432	11,5	13,00	47	9
2005	4368	11,5	9,75	46	10
2005	4368	11,5	11,30	43	9

Thus we see that the average time required for the complete turning of a locomotive wheel center is only 54.6 minutes, including the time spent on setting up.

This lathe requires up to 29 HP and is designed for 35 HP. On another IV-Gothic lathe, shown in Figs. 14 and 15, the turning of a car-wheel center was carried out in 52 minutes (11 wheel centers in 9 h. 32 m.), including the time for setting up. The cutters and the turning process were the same as in the first case; photographs of the lathe, of the cutters, and diagrams of the work are shown in the appendix to this article (Plates VI–IX).

The boldness with which American engineers carry out one or another idea, as in the present case the increase in productivity of the wheel lathe, is truly worthy of respect. If among the large domestic works Felzer & Co. in Riga...

build machines of 15 HP, which are considered the most powerful among us, while in America the syndicate of machine-building works Niles-Bement-Pond Co. has already installed a great many wheel lathes of 35–50 HP capacity.

It would, however, be unfair to accuse our factories of a lack of initiative and of conservatism, since the fundamental causes of this conservatism lie in the general organization of our technical life, of which I shall speak below.

I.V.

Let us now examine the entire wheel lathe in full, in its historical sequence, in order to obtain a complete conception of it. For this it is necessary to supplement its diagrams and the descriptions of the principal parts given above with those details which are inherent to it and more or less characteristic of it.

Adhering to the accepted order of the diagrams, we shall begin the description of the machines with the second diagram.

I. The machine shown in Plate 1, as was already said when considering the headstocks and the carriages, is one of the first European designs of the 2nd diagram, as may be judged at least from the fact that Hartmann, who gave a description of this machine, states that it was not widely distributed and proves the superiority of this machine over wheel machines of the 1st diagram.

The bed A of this machine is cast-iron, solid, having the form of a trough with a longitudinal ribbed partition (Fig. 19) and reinforced with transverse ribs (Fig. 17). To the longitudinal rib is attached a toothed rack u, along which run the gears of the carriage cast-iron boards P, moving these foundation boards together with the carriages along the bed A.

With the aid of this same rack, by the gearing of the gear t (Fig. 19), the movable headstock J is moved. The spindle shafts are cast-iron, and the distinctive feature of the shaft of the fixed headstock is the permanent, non-projecting center l (Fig. 20).

The movement of the bed Q of the carriage perpendicular to the axis of the half-slide is accomplished by simple advancement along the carriage board P; for this it is necessary first to loosen the bolts fastening the carriage to the board. These cross carriages in those times (the end of the 1840s) were also a novelty, and therefore in Hartmann we find a description ) of a rather primitive carriage used at that time for wheel lathes, shown in Fig. 27*.

On the board P is mounted a column V with a cylindrical bore, into which enters the clamping lever T with a cylindrical pin.

*) Carl Hartmann. See the appendix of literary sources.

The cutter is placed from above between V and T and is pressed by the screw X against the finger of the holder T, so that the latter does not move away under the pressure of the cutter during work.

For the rapid running of the machine when turning the axle journals, the toothed wheel G is disengaged, d and E are brought into mesh, the wheel E, which rotates freely on the spindle shaft, having first been connected with the stepped pulley D. Thus G or d transmits rotation to d₁ (Fig. 18), which in turn transmits motion through the toothed wheel L mounted on the shaft to the wheel K₁. The feeding of the slide rests, along and perpendicular to the axis of the wheelset, is done by hand.

This machine is a two-slide-rest one, and the turning of the inner and outer lateral surfaces of the tire is carried out by the same slide rests, owing to the possibility of turning the lower carriage through 90°, as shown in Fig. 16 for the slide rest of the left headstock.

In his book Hartmann mentions that, for powerful machines, an internal coupling of the faceplates is made with gears receiving rotation from the shaft; unfortunately, he does not give even a schematic drawing of this construction.

Passing to a general assessment of this machine, it may be said that almost its only defect is the fixed center of the stationary headstock B and its own dead center, of which I have already spoken. The bed A is of the same width both between the headstocks and at the headstocks, although the latter is merely an uneconomical expenditure of cast iron. In modern beds a recess is made according to the type already considered (p. 23).

The wheelset drive is also non-constructive: the drivers were inserted by pins into corresponding sockets of the faceplate, and these sockets became worn out when rotation was transmitted to the wheelset.

On the spindle shaft of the stationary headstock, grooves e are made at the places where the bushings of the stepped pulley are fitted, “to reduce friction,” as the designer explains; but this, of course, is a naive solution to the question of reducing friction, because it entails only a weakening of the shaft; and therefore it would have been much simpler to make the corresponding recess in the pulley bushings.

II. The higher form, in the sense of the design of the headstocks, bed, and slide rests, is represented by the machine described by Hart in 1867. The machine, shown in Plate II, was intended for turning the tires of wagon and tender wheelsets and was manufactured by the R. Hartmann factory in Chemnitz.

First of all, attention should be paid to the centering of the wheelset. Here this is done with the aid of bushings m (Fig. 7), bored on a taper, into which the axle journals of the wheelset enter with

fitted on them with shoes (each of two halves), likewise turned to a cone. By advancing the right movable headstock \(b_1\) with the aid of a screw, set in motion by the handle \(S\) and the gears \(t\) and \(u\) (Fig. 2), the journals of the wheelset, with the shoes fitted on them, enter the bushings \(m\) and are clamped by these shoes. When the journals sit in the bushings firmly enough, the wheelset is additionally supported by the centers of the spindle shaft, in order to remove part of the pressure from the bearings of the spindle shaft and transfer it to the dead centers \(Z\) and \(Z_1\). The construction of this headstock is important for us in that it serves as the first embodiment of the idea of self-centering of the wheelset, which among American engineers was subsequently realized in the machine already considered by us, with self-centering lunettes (p. 17, Fig. 11).

No less interesting in this machine is the transmission of motion to the faceplates, which may likewise be regarded as the prototype of the American design of a machine of the fourth scheme. Here, first of all, the stepped pulley is carried outside and frees the bearings of the spindle shaft of the movable headstock from the load arising from the overhead drive.

On the shaft \(d_1\) of the stepped pulley \(d\), passed through the frame \(a\), there is mounted a gear \(e\), from which rotation is transmitted by the gear \(f\) to the second shaft \(f_1\), with gears \(g\) and \(g_1\) (Fig. 9), transmitting motion to the gear wheels \(h\) and \(h_1\) (Fig. 2), seated on the shafts \(i\)–\(i_1\). Finally, the gears \(k\) and \(k_1\) of the shafts \(i\)–\(i_1\) transmit rotation to the toothed wheels \(l\)–\(l_1\) of the faceplates, to which the bushings \(m\)–\(m_1\) are fastened by four bolts.

It may be said that the shaft \(f_1\), with the gear wheel \(f\), in the middle recalls the fourth scheme of American machines for transmitting rotation to the faceplates; it is clear that here the transmission of motion to the faceplates \(l\)–\(l_1\) is very complicated, since the triple transmission is extended over two additional shafts; nevertheless such a transmission scheme undoubtedly frees the left headstock as well, which is usually immovable. True, here too the left headstock is immovable, although for convenience of centering it ought to have been made movable; but this may be regarded as an unfinished idea of the designer.

The right headstock moves along the frame together with the shaft \(i_1\) (Fig. 2), while the bearings of the gears \(g_1\) and \(h_1\) are immovable; both gears remain in place, the gear \(h_1\), being freely mounted on the shaft \(i_1\), allowing through its bushing the keyed portion of this shaft to pass. The fixed left headstock differs from the movable one only in that here there is no feed screw and the gear \(h\) is mounted fast on the shaft \(i\).

Let us now consider the device for automatically moving the cutter in two mutually perpendicular directions. In the preceding machine the carriages of the rests, as we have seen, were moved by hand. Here, however, by means of a ratchet device and a counterweight, the rest is given these movements automatically, in the following manner: to the lever of the counterweight is attached a chain, thrown over two guide pulleys above the machine; by its other end the chain (Fig. 1) is connected with the crank of the gear wheel r, rotating from q, which is mounted on the spindle shaft; a section of the crank gear is shown in (Fig. 8). In this way, during the running of the machine, the screws p and p₁, or o and o₁, turned by the ratchet system, feed one or the other carriage of the rest, according to the character of the work—whether it is the turning of the tread profile or of the side surfaces of the tire. This system of feed is still used even now, with the same arrangement of the gears q and r, sometimes carried out onto the tail end of the spindle shaft. By shifting the crank of the gear r and changing the length of the arm of the lever with the counterweight, one can increase or decrease the amount of rotation of the screw, and consequently the feed of the cutter.

The bed of this machine, in its construction, is in no respect inferior to the best modern design: it is solid and strengthened with transverse U-shaped ribs. The beds of the headstocks are hollow (Figs. 10 and 11), with split bearings; moreover, the bed of the right movable headstock (Fig. 11) has a boss for setting in it the nut connected with the screw that moves this headstock.

III. The milling lathe, although it differs sharply from the other types in its working tool, nevertheless, in the form in which it was built by Schiess in Germany and by the Bromley Brothers here, may be assigned to the second scheme, according to the construction of the headstocks and the mounting of the wheelset.

In Fig. 31 is shown the milling lathe of the Bromley Brothers works in Moscow. The rests of this machine have already been described on pp. 24–25, and therefore we shall dwell only on the machine itself.

First of all, let us note that the machine is driven by the stepped pulley B, receiving motion from the ceiling drive. The axis of pulley B is connected with the worm a located on it, and with the worm wheel K; in turn, the axis K, passing through the stationary headstock, carries a screw gearing with the main shaft v, on which are mounted the gears that set the faceplates in motion. By means of the levers r, these gears can be disengaged from coupling with the faceplates. The right movable headstock is moved by means of the screw M, entering into the self-acting nut of the bed. For moving the rests there serve the screws w and w₁, with square ends of their ...

The Wheel Lathe

Fig. 31.

the slot m, as does M of the movable headstock, for inserting a wedge: when the supports are shifted, the bolts b are loosened, and the support board can move freely.

This drawing also shows the steps for supporting the spindles in the event that locomotive half-axles with crank pins have to be machined and the thrust center is extended far out.

In addition, since practice, as I have already said earlier, has set limits to the unthinking fascination with the milling cutter—to which the milling machine must also be assigned—this machine, owing to the complexity and high cost of its transmissions (screw gears) and its two additional overhead drives, is less convenient than an ordinary wheel lathe; and solely because of these entirely unnecessary design complications it ought to be discarded, although the cutter does indeed give good output.

IV. The machine shown in Plate XI, Fig. 1, the so-called “Rapid,” is manufactured by the former Felzer Works in Riga.

The machine described has a 10 HP electric motor, but there are also many installations driven by overhead drives. The machine is designed with a sufficient margin of strength and is calculated for work with high-speed steel (self-hardening). The fixed headstock (left) carries a spindle shaft with an excellent spindle construction. On the ordinary faceplate there may also be mounted a “steady” Q, playing the role of a supporting steady rest, the upper part of which is held by permanent hinge bolts q; but although this “steady” rotates together with the faceplate and therefore does not scratch the axle of the half-wheelset, nevertheless this device is superfluous and is usually discarded. The external shaft V, passing beneath the headstocks and carrying gears for transmitting rotation to the faceplates, receives its motion in the following manner: from the motor, making 965 revolutions, rotation is transmitted by gearing A and B to gear I, from I to II and III, seated on one shaft with II; then to IV and a on the same shaft; and finally from a motion is transmitted to the toothed wheel b of shaft V. Thanks to the sleeve N, shifted by the handle r, other gears can be brought into mesh, thus changing the rotational speeds. The Felzer machine gives the following speeds: 2—to 3.5 and to 14 meters per minute.

The supports, each with two cutters for turning the profile of the flange, are automatic, of the type described on pp. 29–30; movement of the supports is possible along and across the bed by means of a rack and gear with a ratchet device. The second pair

…plain tool-rests, also with 2 cutters each, is intended for turning the side surfaces of the tire. Both pairs of tool-rests are set in motion by a system of lever transmissions from the crank wheel for the second tool-rests and from the wheel crank for the first. The wheel and the pinion are mounted on the same shaft, the pinion receiving its rotation from a gear seated on shaft V at the movable headstock (the right-hand one), which is moved along the bed by means of screw S. It must be said that this machine is, in our opinion, one of the best, if not the only, designs of the second type, owing to the fact that shaft V is brought out from under the headstock, and all gear engagements are easily accessible, just as all moving parts can be observed during operation. Its only drawback is that the bed is not one piece, and the portion of it beneath the plain tool-rests is connected with the main bed by bolts. This drawback is not justified by any considerations, since casting a one-piece bed is neither large nor difficult, and the joined parts must in any case be planed, so that weight plays no role here in installation.

V. The strongest machines of the second type are nevertheless produced by American works, although in the last ten years the English,*) copying American designs and methods of production, have also begun to turn out fairly powerful wheel lathes—up to 40 HP (among the Americans—up to 50 HP).

Among the American machines of this type I shall note the design of the Bement, Niles Co. works (Philadelphia, a member of the Niles-Bement-Bond Co. syndicate). This machine (Fig. 2) for turning locomotive wheel centers is designed for a maximum wheel-center diameter of 2337 mm (92 inches), and a maximum distance between centers of 2743 mm (9 feet). The right-hand movable headstock is shifted by means of a rack and pinion, driven by a wrench with a ratchet device. At the left-hand headstock there is a variable-speed motor, from which, by means of a system of gear transmissions, rotation is transmitted to the shaft of the machine and from it to the faceplates. In addition to changing speeds by the motor, two further speeds can be obtained by means of clutch T, which engages either the right-hand or the left-hand system of gears.

VI. The machine of the Niles Tool Works (Hamilton, Ohio), from the same syndicate, shown in Plate XI, Fig. 3, takes up to 50 HP. This machine is designed for turning locomotive wheel centers with diameters from 1321 to 1727 mm (52–68″); the greatest distance between centers is 2743. It is driven by a variable-speed motor, the speeds also being regulated by different gear-wheel engagements, which makes it possible to vary the speeds from 3.05 to 9.14 meters per minute. The latter speed is even suitable—

*) The works of Loudon Brothers, Ltd., London.

for turning the necks. In addition, the electro-magnetic clutch \(S\)--\(S\) makes it possible instantaneously to reduce the speed of rotation to \(102^{m}/m\). The faceplates with cut-outs for receiving the crank pins have internal gearing. The massive bed is of cast iron; the rests are simple, set in motion by a lever system. The right-hand movable headstock is moved along the bed by means of a separate motor.

A characteristic feature of American machines driven by electric motors is that their speeds of rotation are changed directly by the motors as well; that is, in other words, the Americans use almost exclusively variable-speed motors, greatly reducing the bulk of gear transmissions.

In concluding the survey of machines of the second type, one should also note one of the special appliances for machines for turning locomotive driving-wheelsets: lunettes, installed on the bed of the machine to support the axle of the wheelset. They are used only by Russian and English works. Practice, however, rejects lunettes even with our primitive method of fastening wheelsets; if one passes to the American method of fastening, then, naturally, lunettes will prove to be a technical anachronism. Another appliance, practiced almost exclusively by American works, is the device for boring, in the wheels, the holes for the crank pins. However, this incidental American enthusiasm for the universality of the wheel lathe is applied rarely, and the very works that build machines with a boring attachment recommend special boring machines for this purpose.

Having finished with machines of the second type, let us proceed to the designs of the third type.

VII. Examples of machines of the third type, first category, may be the almost entirely identical designs of machines with “bearers,” manufactured by the Moscow works of K. A. Weichelt and the Bromley Brothers. These machines appeared, as I have already said, as early as 1873 at the Vienna World Exhibition, exhibited by the firm R. Hartmann in Chemnitz. In the West they have long since been discarded, but here, owing to their relative cheapness, the above-mentioned works still manufacture them, having altered the almost 40-year-old design very little, apparently honoring the precepts of antiquity most profoundly.

On p. 45 such a machine of the Bromley Brothers works is shown, operating, among other places, in the Alexandrovsk workshops of the Nikolayev Railway. These machines are adapted exclusively for turning wagon and tender wheelsets; therefore the height of the centers does not exceed \(610^{m}/m\). Both faceplates receive motion from

Wheel Lathe.

of the shaft \(V\), rotating in the journals of the headstock frame, as is clearly seen

Fig. 32. Lathe with copper [[unclear: plural noun omitted/obscured]] of the Bromley Brothers works, Moscow.

according to the diagram on p. 7, fig. 5. The gears of shaft V can be disengaged by levers r. Shaft V receives rotation from the stepped pulley by means of the toothed back gear Z—Z; the screw K of the stepped-pulley shaft meshes with the worm wheel v, which has a crank, from which, by means of chains or lever transmissions, the carriages receive automatic travel. As was already said earlier, the “bears,” i.e., the adjustable bearings P, are in practice rejected as not attaining their purpose (fig. 32).

Of the variants of this lathe, an attempt was made—unsuccessfully—to replace the cylindrical gear wheels of the faceplate with internal screw-gearing with rotation transmitted from screws: a wheel-turning lathe was built under the patent of engineer E. C. Pfaff by the Saxonia works. But this lathe did not proceed beyond the Vienna World Exhibition of 1873.

VIII. The unsuccessful European design of the 3rd type was replaced by the Americans with the lathe shown in fig. 33, whose headstock was described in detail on p. 17. This lathe, for wheels of the largest diameter, 1067 m/m (from the Niles Tool Works), receives its motion from a variable-speed motor, and is equipped with two cross carriages receiving automatic feed from a system of levers. By means of the flywheel K the non-rotating spindle is moved, while the small flywheel K₁, similarly to a counter-jack, secures the spindle in the set position. R is a roller with adjustable counterbalanced templates corresponding to the profile of the tire, for checking the turning both by profile and by wheel diameter. In this lathe we see the old system of fastening the half-skate by means of guides. In addition, access is difficult to the gear transmissions that advance the self-centering planets. The latter has been remedied in the same lathe by K. A. Weixelt (Plate III) by carrying the end E of roller D out of gear C to the same place where the small flywheel of the spindle is located, which makes the shifting of the planets accessible and convenient.

Fig. 33.

This, in fact, may conclude the survey of machine tools of all existing types, because the 4th type has already been described by me with sufficient completeness on pp. 20 and 21. A comparative assessment of the separate types was also made, and we came to the conclusion that the 4th type eliminates the shortcomings of all three preceding ones. These shortcomings, however, are substantiated more or less theoretically, or rather their assessment is to a considerable degree theoretical. If, however, we turn to the accounting of their shortcomings, so to speak, in time and space, then it may be said that all three of the last types of machine tools—where the 3rd is the American one—are quite equal in rights in wheel shops. In fact, if a Sharp, Stewart & Co. machine tool in the St. Petersburg shops of the Petersburg–Warsaw Railway could work for 50 years, then it is rather difficult to determine the wear of a machine tool by introducing an independent variable—time. The wear of a machine tool is determined, of course, by observation; but if a machine tool of an old manufacture and old construction works for 50 years, then hardly even two human lifetimes would suffice for observations within the limits of tangibility and scientific validity of the results of these observations. A far more serious question than the wear of one or another construction of machine tool is the question of their coefficient of useful effect. However, the acuteness of this question is considerably lessened if one takes into account the negligible percentage of useful work of heavy machine tools.

The sole requirement that must be imposed on wheel lathes is the requirement of speed, and, as a result of it, high productivity. However vulgar the American saying “time is money” may be, it is applicable also to the evaluation of the design features of machines, because “time is money” is the basis of the modern process of production, in which the technician and the worker play the principal roles. And, remaining faithful to this principle, American engineers created a fully expedient construction of wheel lathes, because the genius of invention, purifying this principle of its vulgarity, strove toward the economization of human effort and of the energy obtained from fuel.

Let us now see how our railways, which are almost the sole consumer of wheel lathes, relate to the new word of technology.

Our roads ought to be richer in comparison with the American ones, because our net profit per verst is 5,000 rubles, whereas among the Americans it is 4,000 rubles, i.e. 20% lower. But we must be both economical and demanding toward technical equipment, which has cost us much more dearly, as is indicated to us by the percentage ratio of net revenue to the cost of construction of railways: among

in our country it is 4.3%, and in America 5%. Yet over 10 years (1894 to 1904) our receipts fell from 4.5% to 4.3%, whereas among the Americans they rose from 3.1% to 5%. The latter figures sufficiently emphasize the decline of our railway economy; but it is also interesting to trace by what facts of technical inexpediency the loss of time and money by our railways is accompanied.

The obsolescence of the equipment of our railway workshops has now already become proverbial. Yet there is no hope for a speedy renewal of this equipment, because first of all one has to contend with the hidebound ignorance of railway engineers, which is being successfully exploited by our factories under the protection of a prohibitive tariff. How characteristic the effort of Russian factories to take account of this ignorance is may be judged from the following extract from competitive bids for the supply of wheel lathes in response to an invitation from one of the best railways of the Petersburg district. In announcing the technical specifications, this railway did not require the most important thing: the productivity and the amount of power needed for the machine. And this is how the factories replied:

FACTORIES:	Phoenix	Felzer	Bryansk	Br. Bromley	Kramatorsk	Weichelt	Gerlach and Pulst	Geisler	Office of San-Galli
Price, rub.	26000	23620	23775	10900	9150	11500	14400	10650	19000
Time of manufacture, months	5	6	9	5	6½	5	6	10	4—5
Weight, poods	2000	2880	2000	1900	1400	1800	2150	1750	1170
Number of HP	—	10	—	—	—	--	10	15	15

In the inquiry a definite model of machine was specified, for turning locomotive wheel pairs, with a height of centers of 1100 mm and the greatest distance between them of 2800 mm; automatic tool slides were required for turning the tire profile, and simple tool slides for turning its side surfaces.

In the factories’ replies, what is most characteristic first of all is the fluctuation in weight—1170 and 2880 poods. The weight alone already indicates the different power of the machines and, consequently, their different productivity. Therefore the factories prudently kept silent about the number of HP, except for Felzer, Gerlach and Pulst, Geisler, and the office of O. San-Galli, which sent a description of a Niles-Bement-Pond Co. machine requiring 35 HP and, evidently out of modesty, specified 15 HP. But how else can a factory or technical office treat a railway, which in tech-

technical specifications require that faceplates be mounted on “strong” spindles. It is clear that, if such a sonorous and empty term from an advertising catalogue is admitted into the technical specifications of a railway, then the factories also answer with equal frivolity, offering machines at a price of from 5.8 to 16 rubles per pood, i.e., with a triple difference in price. Of all the factories, only the former Felzer & Co. in Riga presented fully detailed data and a non-advertising description of its “Rapid” machine.

Be that as it may, new and powerful designs of wheel lathes are making their way even into the archaeological museum of our railway workshops. Nevertheless, the foreman of the wheel shop—usually an engineer, overwhelmed and absorbed by the performance of his bureaucratic duties, as in every other shop—almost never takes an interest in the technical side of the matter. The machine falls under the care of an ignorant gang boss, who begins to mutilate it in accordance with his own technical inspiration and with the workers’ inability to deal with the new machine. And as a result, one obtains, for example, the following picture: in the Aleksandrov Workshops of the Nikolaev Railway the automatic tool carriages of Gerlach and Pulst were discarded because the workers, gang bosses, and “the engineers themselves” found them unsuitable. These carriages were replaced by simple cross-slides. But in the Petersburg Workshops of the Warsaw Railway these carriages gave almost twice the productivity of the simple ones, and this depended on the fact that the foreman of the wheel shop knew how to teach his workers to handle these carriages.* An even more striking example of engineering thoughtlessness is presented by the Putilov Works, where, on the wheel lathes of the locomotive-mechanical shop, these same Gerlach and Pulst carriages were rejected, while in the wheel shop of the wagon workshops they were very much liked. It may be assumed that at the works this phenomenon was accidental, although, in view of the general course of our factory management, this assumption is made with great hesitation; but in railway workshops our technical ignorance and incompetence are an established fact. This is not surprising, because the “rank” of shop foreman, for a railway engineer, is only one of the lower rungs of the hierarchical ladder leading to a higher salary. Under such conditions, of course, there will be little room for technology, and “the engineers themselves,” in evaluating one or another machine, will always be guided by the advertising of the catalogue and by the ignorant expert opinion of a worker or gang boss.

* I take this opportunity to express my deep gratitude to the wheel-shop foreman A. N. Noman for the help he gave me in collecting practical data for this article.

I shall try to make a comparative assessment of our and American work in turning wheelsets, from the technical and economic point of view.

Despite the fact that in the United States of North America no more than 10% of car wheelsets have tires, while the remaining 90% are cast from iron with a hardened running surface and, consequently, are not re-turned when worn but are sent for remelting, the Americans strive to reduce even this, comparatively small in relation to ours, quantity of turning work after rolling and of re-turning wheelsets after wear. They have, for example, none of such wholly unproductive work as turning the side surfaces of a tire, which is positively justified by nothing. But the amount of turning work is further reduced there by the rational use of the air brake, owing to which they have almost none of those phenomena of wheel sliding on the rail which cause flat spots, just as there are no flat spots from brake shoes. Thus re-turning is required only because of natural wear.

Thanks to rationally conducted tests of wheel lathes, the Americans are developing a type of the highest productivity; they eliminate superfluous operations, such as turning the side surfaces of the tire, and in this way reduce the cost of turning a wheelset by at least 100% as compared with ours.

Let us see what the conditions are for producing this work in Russia. First of all, on our railways there are not only no tests of lathes for productivity, but there is not even any discussion of demanding a definite productivity from a lathe of one firm or another. The quantity of re-turning work is extraordinarily large, because the wear of tires on these roads is almost twice as rapid as on German or American ones. By way of illustration of the latter, one may point at least to our Siberian express-train sets, observations of which have shown that a wheelset runs no more than 20,000 versts without re-turning, i.e., in other words, only two one-way journeys from Moscow to Irkutsk. In addition, poor assembly of air brakes, inability to handle them, and inspection of them by inspectors who poorly understand the matter give an enormous number of flat spots in the tires. As a result, the service life of our tires fluctuates from 3 to 8 years, if turning is performed according to ministerial rules, i.e., if wear (rolling-out) of no more than \(5^{m}/m\) is allowed for passenger wheelsets and \(7^{m}/m\) for freight wheelsets, whereas Krupp tires last twice as long.

One way or another, we spend on these operations no less than 100% more than the Americans, not to mention the loss of time and the initial costs of the increased shop floor area ma-

workshops when installing low-productivity machines. And this applies not only to wheel-turning work, but also to all other work, still less studied than wheel work.

In concluding this monograph, I must note that I by no means intend to prescribe, in the form of a recipe, one or another wheel lathe—whether in the sense of a patent or of a factory—as a remedy for all the woes of our wheel shops. The aim of this work has been a scientifically objective study of the design of one of the most expensive machine tools in our railway workshops. And, of course, it is not for me to judge how successful this study has been; but I am deeply convinced that work in this direction must be continued, because the rational equipment of workshops is an urgent question, for the solution of which we as yet have no material.

P. A. Kozmin.

Literary Sources

1) Albums of machine tools for machining metals. Compiled by Eng. Vasiliev. Library of the St. Petersburg Technological Institute.

2) Wilmuth, mechanical engineer.—“Wear of steam-locomotive wheel tires.” Engineer, 1885.

3) A. Gatsuk, Prof.
a) American machine tools for machining metals. St. Petersburg, 1896.
b) Lithographed course in the mechanical technology of metals. St. Petersburg, 1905.

4) Gadolin, Prof. and Academician. Mechanical technology. Lithographed course. St. Petersburg Technological Institute, 1888/9.

5) K. A. Zvorykin, Prof. “Work and force required for separating metal chips.” Moscow, 1893.

6) V. Knabbe, Prof.
a) Modern equipment of machine-building plants and railway workshops. Kharkov, 1896.
b) Machine tools for cold working of metals. St. Petersburg, 1902.

7) N. Savvin.—High-speed steel. News of the St. Petersburg Polytechnic Institute, 1905.

8) Reports of the railways for 1905.

9) “Information on the Railways”, 1904. Publication of the “Statistical Department of the Ministry of Railways.”

10) Ernst A. v. Hesse, Maschinen-ingenieur in Wien. Die Werkzeugmaschinen zur Metall- und Holzbearbeitung nach den Ergebnissen der Wiener Weltausstellung. Leipzig, 1874.

11) “Engineer”, 1906, Dec. 28.—Test of an American lathe of the Niles-Bement Pond Co. system.

12) Dr. Carl Hartmann.—Vollständiges Handbuch der Metalldreherei.—Weimar,

13) J. Hart, professor of mechanical engineering at the Polytechnic School in Carlsruhe.—Die Werkzeugmaschinen für den Maschinenbau zur Metall- und Holzbearbeitung.—Heidelberg, 1867, and 2nd ed. 1874.

14) Ledebour A., Prof. of the Mining Academy in Freiberg.—Die Verarbeitung der Metalle auf mechanischem Wege.—Braunschweig, 1877.

15) Prof. J. T. Nicolson and Mr. Dempster Smith.—Engineer, 1906, Dec. 14.

16) “Progress Reporter.”—A remarkable record in the turning.—New York, March, 1906.

17) W. Taylor.—The shape and duty of roughing tools.—American Machinist, March 2—9, 1907.

Plate I.

Fig. 17.

Fig. 18.

Fig. 19.

Fig. 20.

Fig. 21.

Fig. 22.

Fig. 23.

Fig. 24.

Fig. 25.

Fig. 26.

Fig. 27.

Plate II.

Plate III.

Fig. 1.

Fig. 2.

American lathe of the A. Beichent Works, Moscow.

Plate IV.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Plate V.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.

Fig. 7.

Fig. 8.

Fig. 9.

Fig. 10.

Plate VI.

Lathe of the Niles-Bement & Pond Co., tested in 1905 at the workshops of the New York Central Railroad Co.

Plate VII.

3 revolutions  11 revolutions

0 revolutions  1 revolution

Line of wear.

Line of rough turning

fig. 1.

Roughing pass of the cutter in 16 revolutions.

2 revolutions  2 revolutions  2 revolutions  2 revolutions

fig. 2.

Finishing pass with four types of forming cutters.

Plate VIII.

Band after 11 revolutions of rough turning of the running surface.

Plate IX.

Rough-turning chips.

A cutter during finish turning of the running surface.

Plate X.

Profile cutter for finish turning the inner part of the flange.

Profile cutter for finish turning the outer part of the flange.

Plate XI.

Fig. 1.

Fig. 2.

Fig. 3.