Operation of Parallel and Radial Axies of a
Locomotive by a Single Set of Cylinders.
By Anatole Mallet.
The American Society of Mechanical Engineers, 1915
An examination of the principal arrangements proposed for transmitting power to convergent axles without, however, increasing the number of steam cylinders forms the subject of the present paper.
These various sytems of transmission may be divided
into two classes :First, those which involve elements having
rotary motion ; sccond, those which involve elements having reciprocating motion.
Transmision by Rotary Motion.
This class includes gear transmission, transmission by endless
chain and transmission by universal joints.
Gear Transmussions. Although the first use of gear transmis-
sions dates back to the very origin of locomotives, they appar to
have been first utilizied for operating locomotive axles having freedom of radial movement in 1838, in a locomotive built at Heath Abbey for the Rhymney foundry in Wales.
This locomotive was carried on two trucks with two axles each. The two trucks could turn so that they were at an angle with each other without throwing the driving gears out of mesh.
In I841 the Baldwin Locomotive Works built a locomotive in
which the rear axles were driven by means of a countershaft and connecting rods, and the axles of the front truck were operated by a gear transmission located on the longitudinal axlis of tbe machine.
This locomotive weighed 13 tons and was designed for use on a
quarry railroad, but the type was afterwards abandoned.
Fig1. Six-axle Gear Transmission locomotive of Tourasse
The French engineer. Tourasse. presented at the Competition
of Semmering in 1851 a design of locomotive with six axles, similar to the Rhymney locomotive. It may be seen from fig.1 that from the cylinders is operated a countershaft carrying a toothed gear which engages with toothed wheels carried on the nearest axle of each track, these axles in turn being coupled to the other axles by outside connecting rods. This locomotive was to weigh 60 tons with the water carried in a saddle tank in the boiler.
The cylinders were 0.50 m. (1.64 ft.) in diameter with a 0.60m. (1.97ft. ) stroke; the wheels were 1.20 m. (3.93 ft. ) in diameter and the heating surface 250 sq. m. (2,590 sq. ft. ). The power developed would have been extraordinarily large for that time, since, according to the author of the design, the locomotive was to be able to start with a lord of 250 tons over a grade of 2.5 per cent., although a capacity of only 140 tons was required.
The Locomotive Works of Winterthur, Switzerland, bullt in
1883, for an Industrial railroad in the south of France, a locomotive similar to the one just described. It was supported on two trucks of two axles each, and weighed 22 tons. It appears that this type was unsuccessful.
The famous Engerth locomotive (fig. 2), built after the Semmering Competition from which no practical results where obtained, was at first characterised by the use of gear transmission for connecting the last axle of the locomotive to the forward axle of the tender. The intermediate shaft, carrying the middle toothed gear, is arranged to slide longitudinally in its bearings if necessary, to cut the connection with the wheels of the tender.
Fig.2 Original Engerth locomotive.
Quite a large number of Engerth locomotives were built. As
a rule they had six driving wheels under the locomotive proper
and four under the tender, or a total of ten in all. As the gears
did not give satisfactory results in actual practice, however, they were eliminated and the machine reduced to the type of locomotive and tender with six coupled wheels. The complicated gear transmission type has long since entirely disappeared from practice.
Within recent years a locomotive builder in Lyons has built
some small narrow gauge locomotives which are supported on four axles-all driving. The three rear axles are coupled by external connecting rods, while the front axle, which has radial freedom of motion, is connected with the axle next to it by a train of gear wheels located in the longitudinal axis of the machine, just as in the Engerth type. It does not appear, however, that this system has found an extensive application.
Before 1830 W. N. James, of Birmingham, proposed to connect not only the axles of the locomotive and tender but also those
of the cars by means of gear wheels operated bv a longitudinal
shaft running the length of the train and provided with ball and
socket joints to give them the fiexibility necessary for making the curves. By this arrangement the inventor proposed to obtain
sufficient adhesion to handle the train on grades without recourse to the Blenkinsop rack, and he stated that experiments made on a small scale showed that he could make grades of three inches in a yard, or 1 in 12.
There are, in actual use in the United States, Iocomotives in
which the axles of the engine and of the tender are coupled together by gears and a longitudinal shaft fitted with ball sockets. These locomotives are of the Climax, Shay and Heisler systems, which differ from one another in the arrangement of the details of transmission and the location of the steam cylinders. These systems are to well known to necessitate their description here, but it may be of interest to state that they have been used even for very large units. Thus, the Climax and Heisler locomotives have acutally been built in sizes of 75 to 80 tons and the Shay locomotives up to 135 tons.
Transmissiom bv Endless Chain. The use of endless chain for
coupling axles which may be thrown out of parallelism appears to have been adopted for the first time in 1851 by S. A. Maffei, of
Fig. 3. Locomotive " Bavaria," by Maffei.
Munich, in the construction of the locomotive Bavaria, presented at the Semmering Competition. This machine (fig. 3) had seven axles, driven by two cylinders. The axles were divided into three groups and the wheels of each group were coupled by external connecting rods, while the groups were connected by endless chains made of links and studs. The engine weighed 68,000 kg. (74.8 tons) with its equipment and it was given the first prize at the Competition after having satisfactorily passed all the tests.
It was said that the victory was due only to the very brief duration of the tests, and that this locomotive could be maintained in good operating condition only by constant repairs to the chain transmissions. As a matter of fact, the Bavaria has never been reproduced in full or in part.
The Locomotive Works of Winterthur built a locomotive with
three axles, the middle one rigid and the two others with radial
displacement. The two non-rigid axles were connected by Galle
chains (Straight-link roller chain) engaging with rim gear sprockets attached to the axle by ball sockets.
The cylinders were located forward, were vertical
and drove a countershaft which was coupled with the axles by the Galle chains. The locomotives of standard gauge were 16･5 tons and could handle curves of 11m. (36.08 ft.) radius. This type was not built again.
In the United States a type of small locomotive designed to
operate over roads made of logs placed end to end is sometimes used in lumbering operations. This locomotive is set on two trucks and its axles are driven by means of Galle chains from a countershaft operated by the cylinder. The wheel treads are grooveshaped, fitting over the log rails on which they run. It is generally known that Galle chains are used on road locomotives, rollers, gasoline locomotives and certain electric locomotives.
Transmission by Univefsal Joints. In this category can be placed all transmissions by ball joints. The term ball joint applies here to any device involving wheels mounted on a hollow axle, in the interior of which is a shaft that receives the power from the
steam cylinders and transmits it by means of a ball joint, or universal joint, to the hollow axle. On curves the hollow axles take the radial displacement while the interior shafts remain parallel.
The interior shafts are coupled by external rods in such a manner as to make the converging movements of the outer axles and the movement of the transverse displacement of the central axle correspond. This ingenious device appears to have been invented by Percival Haywood, who made use of it about 1880 on a small locomotive weighing approximately 2.5 tons and running on a 15-in.gauge railroad having curves of 16 ft. radius (fig. 4) The three axles are connected.
Fig. 4 Haywood's Hollow-Axle Locomotive
E. P. Cowles applied the same principle, but a different arrangement, to a locomotive on a quarry railroad in Kentucky as shown in fig. 5 (only one-half of the machine is shown, the other half being exactly simillar). From the figure it may be seen that only the central carrying axle of each truck is hollow, containing a rigid shaft a acted upon by the steam cylinders.
The other axles are coupled by external connecting rods, in the middle of each of which is provided a slot in which slides the crank pin of the shaft a. The shaft a is carried on external supports and is connected with the hollow axle by a central universal joint.
The inventor utilized the peculiar idea of operating both trucks
from the same cylinders in order to simplify the general construction of the machine. To accomplish this each piston rod was arranged to pass through both covers of its cylinder and to engage at each end with a connecting rod. Due to the obliquity of these connecting rods, however, their mid points of stroke did not correspond to each other, and a sliding of the wheels on the rails twice in each revolution resulted. This difriculty could have been avoided by the use of two pistons in each cylinder, one for
Fig.5 Cowles' Hollow-Axle Locomotive.
the foward truck and the other for the rear truck, but that would have complicated the machine.
Transmisson by Reciprocating Motion
This class includes such systems as make use of connecting
rods, equalisers, etc., for connection of the convergent axles. The author considers it advisable to call attention in a general manner, however, to the fact that this classification cannot be very rigorous because certain arrangements might belong to two classes at the same time, owing to the multiplicity of parts entering into their construction.
The mechanisms of his class may be divided in the following
manner : Coupling of convergent axles by connecting rods located in the longitudinal axis of the engine, these connecting rods being either simple or double, rectilinear or triangular ; coupling by oscillating levers or equalizers; use of a free axle coupled by connecting rods to the converging axles, and coupling of axles by means of external connecting rods of which the length varies with the radial displacement of the axles.
Coupling by Connecting Rods Located in the Longitudinal Axis
of the Engine. The use of central connecting rods acting on
sperical crank pins located in two continuous axles is a very simple idea, but there is serious difficulty in passing the dead centre. This can be remedied in several ways. In designs pressented for the Semmering Competition, Maffei proposed to locate, side by side and in the centre of the axles, two connecting rods acting at righht angles on cranked portions of the axles. It is easy to see that this solution of the problem was not a rigorous one, since the connecting rods are not on the axis of the engine, although they are very close to it. A certain amount of play had to be allowed rods between the pins of the crarnks and the brasses of the connecting rods, which would finally result in causing a breakdown of the transmission .
Fig. 6 Thouvenot's Locomotive with Rods on Axis
Thouvenot took up this idea of locating rods on the axis of the
engine about 1860. He deflected the connecting rords so as to bring them back into the axis of the engine, as shown in fig. 6. This arrangement does not appear to posses sufficient strength.
C. Aliges, former engineer of the Cail factory in Paris, developed a design for a four-axle locomotive (fig. 7), in which one of the two axles forming the truck was connected with the
Fig.7. Aliges' Four-Axle Locomotive
driving shaft by a central connecting rod and the other with the third axle by a like arrangement.
De Bergues, of Manchester, proposed an arrangement for interconnecting the axles of a locomotive and its tender by means of central coupling rods acting on cranks in these axles, the coupling rods being driven by vertical oscilating levers.
This arrangement embodied two similar systems located very close to the longitudinal axis, with cranks at right angles to the axles.
Coupling by Oscillating Lelvers or Equalizers. The idea of using oscillating levers for coupling convergent axles was first disclosed about 1855 in an invention by Lucien Rarchaert, who
tried very persistently to realise it. This system has been very
favourably reported on, but has never been actually used.
A German engineer, Christian Hagans, of Erfurt, invented an arrangement which he applied at first to small locomotives and after, with some modifications, to large five-axle locom,otives with three fixed axles and two axles forming a truck, as shown in fig.8. The axles of the truck were acted upon by a vertical lever, a, oscillated through the intermediary of.a longitudinal rod
Fig.8 Hgan's Five-Axle Locomotive.
by lever a', oscillated by the piston rod. The upper end of the lever a is connected to the top of an equalizer, b, articulated in the middle and having its lower extremity attached by a distance rod to the rear axle. c. of the truck.
The result of this arrangement is that if, on curves, the axles of the truck are displaced, the lower part of the lever a has a displacement in the same direction and to the same amount, so taking care of the convergence of the axles.
The Hagans system was at first considered quite a success on
the Prussian State Railroads on five-axle coupled Locomotives
weighing 72 tons in service, but it has since been entirely abandoned. As a reason for this was given the fact that the introduction of locomotives with five axles, parallel and coupled by ordinary side rods, has made the complication and expensive maintenance of the Hagans machine unnecessary. However this may be, we may say that this system has probably been supplanted by the Engerth system, no doubt one of the most widely applied arrangernents for operating. converging axles.
Fig. 9 . One-Half of Johnstone System Locomotive .
The Johnstone system, which has been applied on several large duplex locomotives built in the United States for the Central Mexican Railroads, has also some resemblance to the preceding type.
Fig 9 shows one-half of this locomotive, and the other half is entirely similar. The piston rod (or rather rods, since there are three of them, there being two cylinders placed side by sicle) acts on the middle of a lever a, which is vertical when in its normal position. The main cannecting rod is attached to the lower extremity of this lever, while from the upper extremity a short coupling rod connects to the top of equalizer b. This equalizer oscillates about its middle and operates from its lower end a connecting rod to a crank pin set at 180 deg. from the working pin of the counter-crank. From the figure it is seen that the lever a, to which the piston rods are attached, moves always parallel to itself, vertically on straight track and at a slight incline on curves.
Use of Free Ax,les. The use of a free axle coupled by connecting rods with radial axles appears to date back to the Semmering Competition. Maffei there presented several designs in which the axles of locomotives and their tenders were coupled by inclined or triangular connecting rods. A similar design (fig. 10), submitted at the same competition , by a Hannoverian engineer, Kirchweger, shows a locomotive carried on two trucks having two axles each, the coupling of the trucks being effected by an arrangement of this kind.
It may be seen that there is a connection betwin the journal boxes of the whheel axles and of the free axle.
The Austrian engineer, Pius Fink, tried to retain in the Engerth
machine its original property of total adhesive weight by substituting for the gear train an articulated device. He built three loccomtotives, which were in service for several years, but they are not built now.
The locomotive had three axles and the tender
Fig. 10 Free-Axle Locomotive by Kirchweger
two, as in the Engerth machine. Dredge and Stein in England
resorted to the use of a free axle and a central triangular connecting rod for coupling the axles of a locomotive with those of the tender.
The inventor, Rarchaert, to whom reference was made above,
after having abandoned the system of oscillating Ievers, designed an arrangement coming under the present category (fig.11). This was applied on a locomotive with two trucks having two axles each. The cylinders operated a free axle coupled with the carrying axles by a triangular central connecting rod. The pins of the cranks had spherical heads. The locomotive was in service on the railroad between
Fig11. Rachaert's Freee-Axle Deign.
Fougeres and Vitre and from Orleans to Chalons.
It gave goodd results, but at the death of the the inventor experiments with it were discontinued.
The use of a loose axle may be combined with that of external
coupling rods in which are provided slots, and in these slots, again, glide the coupling pins of the radial axles. An arrangement of coupling rods of this tvpe was used in the Cowles system, mentioned above (compare fig. 5). The Kochy system (fig.12) also furnishes an example of this arrangement. A third example is found in the design presented by the engineer Gouin, author
Fig.12 Kochy Free-Axle System
of the system employing oscillating levers similar to the first system of Rachaert. This model represents a combined locomotive and tender with five axles, of which the two rear ones form a pivoted truck. These axles are coupled by connecting rods having an open or slotted link in which slides the pin of a crank forming the terminal of a free axle.
This cranked free axle is in turn coupled with the third axle of the machine.
Finally, the well-known designer, Krauss, of Munich, proposed,
in 1893, an arrangement permting of the operation of the axles
of a truck by steam cylinders carried on the main frame of the
engine, as shown in fig.13. To accomplish this, the crank pins on
Figr. 13. Krauss' Slotted Rod Arrangenlent.
the driving shaft carry pin blocks working in slots in the trussed
connecting rods. It may be seen that the use of such connecting rods with slots is subject to serious objections. Stress is exerted on the crank pin in a vertical direction only and, moreover, the pin blocks have on curves a periodic displacement in a direction transverse to the axis of the connecting rod. As a result of this, friction and considerable wear is likely to ensue, rapidly producing play shocks and dislocation of parts. None of these systems appear to have been utillzed practicaliy.
External Conecting Rods, the Lengths of which Vary with
the Convergence of the Axle. In this connection reference will be made first to the Klose system, which has been fairly widely applied. Fig.14 gives an idea of this system. It may be seen from this figure that the crank pin of the working axle carries a kind of rocker lever to two points of which are connected the coupling rods of the other axles. The other two points are connected to the extreme axles by a system of connecting rods and triangles in such a manner that the convergence of the axles corresponds to the
variation in length of the coupling rods. This system has been
employed in locomotives having a gauge of 1.76 m. (5.77 ft.) on the Bosnian-Herzegovinian Railroads and on large five-axle locomotives of the Wurtemberg State Railroads.
An arrangement proposed by a Brazilian engineer, G. Fretl,
might also be cited. In this a double horizontal box engages with
the crank pin of the middle axle and, by bearing on the coupling
Fig. 14･ Klose Variable Length Rod System.
rods of the two other axles, increases or decreases the length of these coupling rods when the middle axle is transversely displaced on a curve.
Conclusion
In this paper the author has indicated the most interesting
arrangements, so far as he knows, which have been proposed for operating the converging and parallel axes on a locomotive way a single pair of steam cylinders. It he has failed to mention any, especially those of American origin, it has been done unintentionally and he apologises for it in advance.
An examination of these devices gives the impression that all
of them involve a serious inconvenience, and that all of them can operate in a satisfactory manner only when they are in vertical play, parallel to the longitudinal axis of the engine, i.e., when the latter runs along straight sections of track. But such is not the condition on curves where the transmission element acquires a certain amount of obliquity, which necessitates the use of pins or spherical parts more difficuit to lubricate.
This obliquity introduces dlfferences in length and further play between the parts, and this, in turn, leads to shocks and rapid wear of parts. Hence the maintenance of mechanisms of this kind becomes necessarily more costly than that of the ordinary locomotive transmissions.
Notwithstanding these difticulties, the author believes that in
view of the ingenuity which has developed in the study of this
question during so many years and by so many inventors, it would be hard to predict that a system may not finally be found combining all the conditions essential to the practical operation of such a device. On the other hand, however, can it not be questioned whether researches in this direction are of any actual utility to-day when there is no hesitation with regard to coupling directly the largest number of parallel axles by external connecting rods and when there are other perfectly satisfactory solutions of the problem based on a different order of ideas?
As a matter of fact, since 1861 and more than fifty years ago,
J. J. Meyer, an eminent engineer and author of the first system
of articulated locomotives which has given practically satisfactory results, wrote the following ;
In the systems proposed for coupling in a rigid manner in what-
soever way it may be, the several axles belonging to two diverging trains, the addition of coupling mechanism introduces a greater complication than the addition of two extra steam cylinders, and the maintenance of these mechanism, as well as keeping the drive wheels rigidly to the same diameter, will be of greater cost than that of the two extra cylinders and the two mechanisms, without taking into consideration the loss in efficiency.
What has been said above is all the more true to-day, since, in
addition to the Meyer machine, we now have the Fairlie system and the author's system, which was the last to come and which has received in recent years such important and remarkable application in the United States, thanks to the energy and skill of American engineers and builders.
ページ製作者注釈
この論文の著作者は、Anatole Mallet氏です。
マレー氏の偉業は言うまでもなく、複式蒸気機関の機関車搭載と複式(半)関節式機関車、マレー機関車の発明ですが、ここでは米国の技術者向けに蒸気機関と動輪の位置関係がfree、な機関車の技術史について解説してます。
実際にはこの後、米国技術者達の討論があるのですが、これは割愛しました。
米国側の意見は、これらは既に過去の技術で、今の蒸気機関車技術のために参考にならない、という意見が大勢でした。
私自身は趣味で過去の蒸気機関車の色々を知るのは好きなので、UPしました。
英語が得意でないので、マレー氏の英文を全部読めたわけではありません。
また、マレー氏の各種機構の説明は、図を見ても未だに理解出来ないものがいくつかあります。これは私が工学の専門教育を経てないからでもあります。
文中、欧州大陸語の独特のフォントが地名などで、一部含まれ、これはUP出来ないので、英文字表記のままの個所がわずかにあります。(例、ウルテンベルグ等)
Hagans機関車と、Klose機関車については、
http://homepage2.nifty.com/kotaroooo/katte.htm上記の本は詳しいです。
ボールドウィンラック機関車
Johnston機関車機構図 蒸気機関車目次 線路を探して目次