On the Frame and Drive Mechanism Designs of Early Steam Locomotives


This paper is a completely rewritten and much expanded version of and replaces the earlier paper, “On the Frame and Engine Configurations of Old Steam Locomotives – In Particular of the ‘Adler’,” by the same author, that had been posted on the web in 2008 and will be cancelled now.

When embarking on a thorough, technical discussion of any given steam locomotive design, a useful approach is to consider the following four major components – each by itself as well as in their interaction with each other: The boiler, the steam engine, the frame/running gear and the drive mechanism. To try to establish an order of importance for those four major components would probably be futile yet, certainly, only the ‘addition of’ the frame/running gear and the drive mechanism make a steam locomotive out of the combination of a boiler and a steam engine.  Because of this fact, it is somewhat astonishing that the frame/running gear and the drive mechanism have received so relatively little attention in the literature on the early decades of steam locomotive development. Furthermore, discussions of the technical issues faced during the evolution of these two major components are even rarer.


The main objective of this paper is, therefore, to help fill this gap with a treatise on these two major components of the steam locomotive that is deliberately of a technical nature. In doing so, special emphasis is put on a brief phase during the evolution of the steam locomotive that was dominated by what is being called here “Stephenson’s arrangement of drive”, abbrev. SA.  At least in the literature available to the author, SA does not appear to have been given  anywhere near the attention it would seem to deserve. In general, the period roughly from 1800 to 1840 is covered. Naturally, the focus is on British locomotives, as most of the early development of the steam locomotive took place in Great Britain.  The North American contributions to said early development are being touched upon as appropriate.


After an introduction, “Abschnitt 1,” a major section of the paper, “Abschnitt 2,” contains brief descriptions and discussions of fifteen locomotive designs from within the above-mentioned 1800 to mid-1840 timeframe. The selection of those fifteen locomotives is, necessarily, based to a large extent on the availability of sources with specific, technical information, incl. technical drawings.  In general, the difficulty of obtaining reliable technical information on these designs needs to be recognized and it has to be acknowledged, in this context, that even somewhat complete sets of original drawings of these early locomotives do not exist anymore.  However, replicas of some of these locomotives have been built over time and the drawings for those replicas were generated and ‘reverse-engineered’ by carefully studying original sources of similar locomotives of that time. Descriptions, in considerable  detail, of some of those activities are available and it becomes clear from those sources that these replicas do not match the respective original locomotives down to the last detail.  Therefore, the insights gained by studying these replicas can not be definitive.


The other major section of this paper, “Abschnitt 3,” summarizes the insights gained in “Abschnitt 2” by providing a succinct, very high level overview of the evolution of the frame/running gear and drive mechanism designs during those approximately first 40 years of the development of the steam locomotive. Some highlights are given here:


Richard Trevithick’s locomotive – the first full-scale working steam locomotive ever built (1804) – relied on its boiler as its structural ‘backbone’ and other locomotives without frames were designed and built until the end of the 1820s.  Yet, starting already towards the end of the first decade of the 19th century, the first designs with a distinct frame emerged. The longitudinal members of those frames, made out of wood, were located inside of the wheels (much like in the case of horse-drawn carriages).


Around the same time, the importance of a somewhat capable spring-system was recognized. This further motivated the use of frames as it is much easier to mount a spring system on a frame than on a boiler. Again following the example of carriages, laminated springs were employed which, because of their internal friction, do not require a separate damping system.


In 1829, the famous Raining Hill contest inaugurated the age of relatively fast locomotives for passenger trains.  The winner, George Stephenson’s “Rocket,” marked the beginning of a new era in locomotive design by replacing the until then common complex gears and levers with a drive mechanism that applied the piston force to the crank axle in a much more direct way. As another novelty, the “Rocket’s” frame was built up out of flat bars.                                

Stephenson’s next locomotive, the “Planet” (1830), mitigated the problem of relatively frequent axle failures by a frame design that had its longitudinal members outside – rather than inside – of the wheels (and cylinders) so that each driving wheel was supported both on its in- and outside. Another cause for derailments was addressed by having the smaller-diameter non-driven wheels forward of the larger diameter driving ones rather than, as in the case of the “Rocket,” aft of them. The “Patentee” (1833), finally, added a pair of non-driven trailing wheels and was built in large numbers.


Despite its considerable success, the above-mentionedStephenson configuration had its weak points, including four intermediate bearings of the driven axle that could move in vertical direction and a rather complex load path between those bearings and the drawing hook as well as the rather complex design of the outside frame and the stretchers. While Stephenson further improved his configuration, his competitors also created new designs :


Already in 1834, George Forrester and Company built locomotives with outside cylinders mounted on outside frames, doing away with the cranks in the driving axle as well as some other undesirable design features.

The first locomotive with an inside frame and its cylinders outside of the frame appears to have been built around 1837 by Eastwick and Harrison Locomotive Works in Philadelphia in the United States.  It has to be noted, though, that its cylinders were mounted on the smoke box rather than on the frame.  The step to mounting the outside cylinders on the inside frame was made in 1846 by the “Licking” locomotive, built by the Rogers Locomotive and Machine Works in New Jersey, again in the United States.


Thus, by the mid 1840s, the search for the overall optimal frame and driving gear configuration had essentially been over: Outside cylinders mounted on inside frames had emerged as the predominant configuration and this wouldn’t change for as long as new steam locomotives were designed.


“Abschnitt 4” of the paper, finally, contains a list of references as well as some comments on the materials used in those early steam locomotive designs and an introduction into the German system of classification of the axle arrangement that is used.