The Friends of the
National Railway Museum

South of England Group
Vice Presidents: Richard Hardy; Sir William McAlpine Bt, FRSE, FCIT, FRSA

Last Update

Talk Synopsis

3 January 2006

Return to Index

Diary of Events

Press Cuttings
Visit Reports
Lecture Reports
Product News


The Engineering and History of Robert Stephenson's Rocket
by Michael Bailey and John P Glithero
12 March 2001

Michael started the talk by pointing out that Rocket has been tucked away in museum since 1862. Until the recent work there was little understanding about the detail of the loco's construction and engineering history. The Science Museum and NRM invited them to investigate Rocket - both a tremendous honour and great fun!

Why them? At the Museum of Science and Industry at Manchester, they headed up the project for the Friends of the Museum to build the Planet Replica. Following this, they were invited to Nova Scotia to examine two ancient engines, Samson - a Hackworth design of 1838 and Albion dating from the1850s. The task was to put these into good order for display. As a result of their investigations, the museum undertook the restoration and conservation of both engines.

On return to the UK they gave a talk at the NRM on their work and pointed to Braddyll at the Timothy Hackworth Museum. It was then in a very delapidated condition, near to being put on the scrap heap. The loco had been outside since 1875, even being used as a snowplough for a while. Dame Margaret Weston was at the talk and invited them to examine the engine and advise on a conservation project. After submitting their report they were invited to undertake the restoration. Braddyll is now displayed inside the Soho shed at Shildon in a colour scheme which has been carefully matched from the paint remains on the engine. A talk on the project was given at the NRM and so came the Rocket project.

The opportunity for the investigation arose because of a two year project to rebuild the transport gallery at the Science Museum. Rocket had to be removed and, as a consequence, was scheduled to spend one year in Japan, and another at the NRM in York. It was during its latter sojourn when the investigation was organised.

Before discussing the investigation, it is worth putting Rocket into its historical context. George Stephenson's Locomotion was slow, cumbersome but did the job of hauling coal and minerals. The requirements of the Liverpool and Manchester Railway were very different , requiring higher speeds and reliability. The Stephensons had to get to grips with a radical new engine design, but the necessary R&D only started when Robert returned for South America in 1827. Lancashire Witch was the first results of these experiments, incorporating the novel idea of sprung driving wheels and inclined cylinders. However he went back to wooden wheels because the available cast iron was too fragile. Rocket followed on and was designed to meet the stringent trial requirements which included targets for haulage and speed. The required speed of 12mph was greatly exceeded; indeed it reached 36mph after the formal trials. Such was the pace of locomotive development, however, that Rocket was out of date almost as soon as the trials were finished. Northumbria, an improved development of Rocket, was produced for the opening of the LMR in Spring 1830, with the Planet class emerging from the Stephenson's factory later that year.

Rocket's History

It was built in Robert Stephenson's factory at Newcastle and sent to Liverpool by ship from Carlisle. After the trials Rocket was extremely popular with the public clamouring for what we would now call "special" trips. However, its working life was interrupted by accidents - the first, of course, being the well-known death of William Huskisson, the MP for Liverpool, during the opening celebrations. The second accident involved a "footplating" railway enthusiast , who fell off the tender following an axle break while Rocket was propelling its train. Rocket was repaired and an opportunity was taken to fit some improvements including a smokebox. Rocket was then put on passenger service - after the opening celebrations it had been relegated to engineering works. Shortly after, Rocket had another accident in Olive Mount cutting, breaking its driving axle. Further improvements were then made. A third accident followed while it was on load to the Wigan branch in 1832, Rocket hitting a loco owned by another company (this was 150 years before TOCs and Railtrack!).

Rocket was then used in experiments, before being sold to the Earl of Carlisle for use in connection with his Westmorland coal mines at Kirkhouse. It only worked here until 1840, when it was put into store, during which time a lot of its non-ferrous parts were removed. An abortive plan to restore the loco for the Great Exhibition was followed by its transfer to the Patent Museum - the predecessor to the Science Museum.

The Investigation.

John then took up the story of the physical investigation of Rocket. These took place in the main hall of the NRM. The sessions were organised for each Monday. During the rest of the week, Michael went round many archives to seek out original documentation, while John wrote up the results of the investigation. During the Monday sessions they felt that they were the main displays at York, fielding many questions from visitors.

Most of the measurements were made with traditional instruments, rules etc., however a novel technique was devised using dentists putty for the measurement of the screw threads of the bolts used on the engine.

The wrought iron bar frame are 1inch thick and, like many other components, have many redundant holes. Invicta of the Canterbury & Whitstable Railway provided an extant model of how the original 38o inclined cylinders of Rocket were fitted to the boiler and frames. When those on Rocket were moved down to 8o, a buffer was inserted at the front, possibly in an attempt to strengthen the frames following the Olive Mount accidents. A major change was necessary to mount the cylinders, via a new sub-frame. Additional buffers for the Earl of Carlisle were also needed to cope with the cauldron wagons. A rough shunt during its time at Kirkhouse was probably the cause of the bend at the front of the frame. There is also a bend at the footplate end caused by interaction with the tender.

The original firebox disappeared while the engine was in store at Kirkhouse. This had a jacketed cover in a horseshoe shape. The backplate was modified to be jacketed probably following one of the earlier accidents. They exact form of the firebox was not known, several inaccurate drawings misleading earlier investigators. There is a footprint on the back of the boiler and there are also clues given by the water level - which had been raised at some time. Sight glass and try-cock holes were blocked off, with new ones introduced, illustrating the raised water level. Originally the level was not much lower than the regulator so, in the absence of a steam-collecting dome, priming was likely; a dome was not fitted until November 1830. The early diagrams misleadingly seem to show a dome, which is now believed to be the front safety valve.

The wheels now fitted are wooden wagon wheels, typical of the time. There is a cast iron crank boss and the wheels are secured to the axle by four keys, using traditional wheelwrights techniques. There are cracks in the spokes which have been secured with a reinforcing collar. The current axle, at 4 ins diameter, is not original - this was only 3½ inch.

The valve gear is a form of slip eccentric and predates Stephenson's link by 10 years. The driving dogs are attached to the axle by pinch bolts. The gear shows several fitting marks put on at different times. One set, comprising a series of dots, can be dated to the 1862 restoration - this is because it has been assembled incorrectly!


The frame is by no means a precision assembly, being some 5/8 inch out of true. This shows the lack of standardisation in engineering at the time. The firebox saddle footprint, in the form of holes on the frame and tube plate, shows that it was also out of true, sticking out by some 1 inch and twisted. This was because the manufacturers were not used to working with copper plate of the thickness required to withstand the boiler pressure. The erection fitters just had to accept the lean and work round the conflict with the cylinder mounting. The holes for steam entry into the boiler backplate are also asymmetric. It also meant that the spacing block feet, which fit the cylinders to the boiler, also had to be individually fitted to ensure the pistons are aligned with the wheels. However on the current Rocket exhibit this alignment is well out of true, especially on the right hand side. This misalignment comes from the 1860s reassembly when the engine arrived at the Patent Museum.

Post-Service Alterations

A deep groove in the axle was caused by the pinch bolts of the slip eccentric, probably originating during the move from Kirkhouse to Newcastle when it was hauled by a locomotive.

In 1862 the engine was in a poor state of repair and the Patent Office Museum asked the Stephenson Works to undertake some restoration. While they managed to get the engine into a presentable form, they also put on some ridiculous additions, e.g. the chimney with petal top, the steam exhaust, a replica tin firebox. The Science Museum curator in 1936 stripped away many of these modifications but added an inappropriate set of trailing wheels and another wrong chimney! More seriously, he discarded the original supplementary buffers fitted for its Kirkhouse days.

The result of all these investigations has allowed Michael and John to produce a diagram showing how Rocket looked when it was first made. Just to show how short a time Rocket remained in this state was illustrated by a report to the owners before the Rainhill trials. The rear trailing wheels had disappeared from the ship when it arrived at Liverpool so a wagon wheel set of 2ft 8½ inch were fitted instead of its original 2ft 6 inch. Rocket consequently went through the trials - and won - with a distinct forward lean!

A report detailing the investigation is available as an NRM publication.


How accurate are the existing replica Rockets? - Three replicas followed the first "Henry Ford" replica. The current replica at the NRM has the wrong firebox, wheels and dome. But in general they are all pretty good representations of the original.

Are the original drawings available? - None are known. Stephenson employed a draughtsman to make assembly drawings. But all of these have disappeared over the last century and a half.

How did the maintenance crew clean the boiler tubes? - With difficulty! This was probably part of reason for retrofitting a smokebox.

Is there any evidence of the early experiments made on Rocket? - A blanked off hole at the top of the tube plate seems to be evidence of the experiment of a rotary engine.

Was any allowance made for the expansion of the boiler as it was heated? - No expansion joints were fitted to allow for temperature or pressure changes but since the temperature and pressure was not very high, it is likely that there were no serious problems.

Was any brakegear fitted to the engine? - No brakes were fitted but the valve-gear eccentric cluster could be moved into reverse while the engine was in forward motion and so produce a counter pressure brake. In September 1830 the Directors travelled to Manchester and on the return a report says that the train was stopped in 70yds for one of the directors to attend to a call of nature! This shows that braking didn't appear to be too much of a problem.