top of page

Testing Locomotives

Stationary testing at the GWR Swindon plant. . . . . . .



GWR 1911 vintage “Saint” class locomotive 2931 “Arlington Court” on the original Swindon plant in the 1920s. Photo: Great Western Trust.

Testing railway locomotive performance has a long history. As far back as 1818, George Stephenson developed a device to measure the pull exerted by the earliest locomotives. In 1839, the Great Western Railway (GWR) were using instruments invented by Charles Babbage which were carried on trains to test their new locomotives’ pulling power and speed. The GWR Locomotive Superintendent Daniel Gooch then went a stage further in 1856 to construct a “measuring van” to travel with trains and take measurements while running. This was an early form of what we now know as the dynamometer car. The GWR’s Chief Mechanical Engineer G J Churchward had built his eight-wheel version in 1901, but a static testing plant was still required. The dynamometer car’s instruments were sometimes used alongside the testing plant to give a greater range of results.


The first rolling plant to test locomotives had been developed in the USA in 1891 at Purdue University, West Lafayette, Indiana. It was this example that influenced Churchward to ask his Board’s permission to build a stationary testing plant at the works in Swindon. This was granted in 1903 and the plant built to Churchward’s design in the locomotive assembly shop was completed the following year. It had five pairs of adjustable and braked four feet diameter wheels on shafts carried in bearings with belts and pulleys for running uncoupled wheels. This first version could test up to between 400 – 500 hp so it was unsuitable for testing larger locomotives operating at maximum power. Excess power produced by the plant was linked to a belt-driven air compressor which helped to power some of the many pneumatic tools used in the works. It started life as a means of testing new and repaired locomotives without needing to take them out on busy running lines, but as time went on and interest in enhancing performance grew, its usage was widened.



The dynamometer car. National Railway Museum

GWR publicity explained in the 1920s how their locomotives could be run at the equivalent of 70 mph with the wheels working at about 400 revolutions a minute while stationary. The dynamometer car could be nearby and connected up so that additional readings could be taken on its range of recording apparatus. The list included drawbar pull, speed, regulator handle opening and cut-off, boiler pressure and regulator pressure, steam chest and cylinder back pressure, exhaust injector pressure, steam chest and smokebox temperatures, smokebox vacuum and gas analysis, water temperatures at several points from tender to boiler, water consumption and coal calorific value and consumption.


In the late 1920s, Churchward’s successor Charles Collett was secretive about his plans to improve the plant’s capacity to handle his latest designs such as the “King” class. A rebuild during the 1930s meant that longer and heavier six-coupled locomotives could now be accommodated. Other refinements allowed for a greater range of measurements. Locomotives could now be tested up to the equivalent of 90 mph, and the GWR Board were once told that 100 mph had been achieved in tests. The plant got the nickname of the “Home Trainer” as it was like static training equipment used by cyclists at the time – an early to mid-20th century exercise bike if you like! It was also a check on the accuracy of wheel balancing and valve setting. Locomotives were brought onto the table or bedplate of the plant using a traverser. The table was then lowered and the plant wheels were adjusted to meet the tread of the wheels of the locomotive under test. Its regulator was opened and it rotated the wheel shafts to which were attached sets of pulleys connected by belt to pulleys below on which resistance was adjustable. Cab instruments showed the footplate crew how to adjust to the running required. The regulator and cut-off were adjusted to represent rising and falling gradients, varying speeds and loads while maintaining steady steam production and consumption.



BR Western Region “Hall” class locomotive 7916 “Mobberley Hall” under test at Swindon in 1951 with the 1901 dynamometer car in the background. Photo: British Railways.

Swindon’s plant was a fine example of continuous improvement. However, it didn’t have the capacity to test all comers. British locomotive testing was enhanced by the use of the new 1933 testing plant at Vitry in France which entertained Nigel Gresley’s 2-8-2 “Cock o’ the North” on its rollers in 1934.



Too big for Swindon! Gresley’s 1934 P2 2-8-2 class “Cock o’ the North”. Photo: British Railways.

This venture was further making the point that a British plant was desperately needed by two more of the “Big Four” companies. Testing was fundamental to locomotive development which was advancing rapidly to achieve greater power and speed. Both the LNER and LMS wanted a modern national testing plant for their purposes. The Southern Railway was more concerned with electrification, although Bulleid’s new “Pacific” classes were still being built from 1941 onwards up to nationalisation. However, for a long time until the Rugby plant opened in 1948, the GWR plant remained the only one of its kind in the country. Depression and the war had intervened, and the second British plant at Rugby didn’t open until 1948, some twenty years after Gresley had advocated more testing capacity. Even with the new facilities at Rugby which tested 37 locomotive types, the Swindon plant which could now test up to 2,000 hp was used for many of the new British Railways Standard locomotive classes. The one-off 101-ton Pacific 71000 “Duke of Gloucester” was one example. Testing helped to get the best from steam traction. Amongst other things, it brought about improved draughting, changes to blast pipes, fitting of double chimneys and enhanced superheating. For Swindon, the works closure in 1986 ended its role in testing.



Mike Peart


Mike Peart is a former railwayman on British Railways (Western Region). He is co-author of Volumes 3 (Freight Marshalling Yards), 4 (Level Crossings) and 5 (Train Detection and Control) of the “History & Development of Railway Signalling in the British Isles” series, and Trains of Hope”, all published by The Friends of the National Railway Museum. He’s been an active Friend of the NRM since 1994 and was one of the four “schoolboy” founder members of the Great Western Society (Didcot Railway Centre) in 1961.

All books can be found on Friends online shop here



335 views0 comments

Recent Posts

See All
bottom of page