Association of Railway Locomotive Engineers designs

These designs were drawn up in 1917-18 at the behest of the British Government.

WHAT MIGHT HAVE BEEN Philip Atkins. Rly Mag., 1990, 136, 164.

Had Britain's railways been nationalised at the end of the first world war, the designs on these pages—proposed by members of the Association of Railway Locomotive Engineers in 1917 and 1918, might have formed the basis of the first standard locomotives.
Towards the end of WW1, when Britain's railways were under direct government control, there was serious talk of nationalising them when peace was restored. Early in 1918 the Board of Trade Journal expressed the view that 'to some extent the material for locomotives has been standardised, but the railways are reluctant to standardise the finished article .... It would seem to be reasonable that the locomotives used in this country should be capable of being reduced to some dozen types'
In fact about nine months earlier the Association of Railway Locomotive Engineers had been requested by the Government to prepare plans for a range of national standard locomotives. Founded in 1890, and composed of the locomotive superintendents of the 30 or so leading British and Irish railways, the ARLE had hitherto been primarily concerned with the possible harmonisation of such facets of locomotive design as tyre profiles. The first details of the ARLE proposed locomotives did not emerge until 30 years later in 1948 when diagrams were published' of Ashford (SE & CR) proposals for a 2-6-0 mixed traffic; and 2-8-0 heavy freight locomotive. These were illustrated again subsequently by Harold Holcroft in his memoirs when he made the following revelations: 'The ARLE decided on a 2-6-0 mixed traffic as a priority ... Only the larger companies were really interested. The most active participants were Hughes (LYR), Gresley (GNR), Fowler (MR), Pickersgill (CR), Churchward (GWR), and Maunsell (SE&CR) ... Though an acceptable scheme for a 2-6-0 was in the making, the method of design by committee, circulation of its draft and approved minutes and drawings was so slow that the war was over long before any working drawings could be available. '
What eventually became of the welter of paperwork associated with this ultimately abortive project is not known, but other evidence survives of the input made by all six railway companies mentioned by Holcroft, plus the Great Central, principally during the latter half of 1917 and early 1918. (Of the larger companies the LNWR and NER were conspicuous by their absence.) This evidence consists principally of the surviving drawing office records (or copies thereof) of Ashford, Derby, Doncaster, Gorton, St Rollox and Swindon, which are now held at the National Railway Museum, York. In 1917 the annual president of the ARLE was G.J. Churchward of the GWR who appears to have exercised a measure of influence, particularly the adoption of a long piston stroke, but certainly not concerning the design of the boilers, over the outline proposals of his own and the other drawing offices. In late July 1917 Swindon DO made a study of H,N. Gresley's GNR 2-6-0 (which had in itself originally been inspired by Churchward's 4301 2-6-0 of 1911), and produced a 'draft scheme for a standard loco'. As Swindon drawing No. 55266 this happily survives on a Derby (MR) glass plate negative which shows an outside- cylinder 2-6-0 with parallel round-top boiler replete with Ramsbottom safety valve, GWR-style cab and tender, outside Walschaerts valve gear, and an outside frame to the leading pony truck. This diagram was first made public in 1954.
The outside-framed pony truck also featured on a rather heavier looking 2-6-0 outlined soon afterwards by the GCR at Gorton, whose diagram was first published in 1951, which has ever since puzzled locomotive historians. In fact J. G. Robinson could have produced a very serviceable 2-6-0 simply by combining the existing 'Director' 4-4-0 boiler with the standard GCR 21 in by 26 in cylinders, but the proposed 2-6-0 was to have 20 in by 30 in cylinders, and a slightly larger boiler. Although the Gorton register logs an associated drawing showing bearing surfaces under the date 17 September 1917, the 2-6-0 diagram itself is dated two months later, on 16 November, when in addition a 'suggested' 4-6-0 goods engine with similar cylinders is listed together with a 'proposed standard' 2-8-0with 21 in by 30 in cylinders, and a 'proposed standard' 4-6-0 engine with four 17 in by 26 in cylinders. As with numerous other intriguing pre-war GC locomotive proposals in the register, these entries are regrettably all deleted in pencil and endorsed either 'destroyed' or 'cancelled'. References continued into early 1918, the last on 5 February somewhat intriguingly stating 'proposed 4-6-0 and 2-8-0 goods engine for 5 ft 6 in gauge.'
Doncaster's proposals were made in August 1917 and consisted simply of Gresley's existing 2-6-0 fast goods engine (with 5 ft 6 in diameter boiler), having the cylinders enlarged from 20 in by 26 in to 20½ in by 28 in, and his 2-8-0 heavy goods with 21 in cylinders enlarged from 28 in to 30 in stroke. In 1917 a piston stroke of 28 in was uncommon, and such of 30 in was unique to the GWR.
In July 1917 the Caledonian Railway schemed a massive 2-6-0, neither of whose 28 in piston stroke and 5 ft 8 in coupled wheel diameter were St Rollox standards. A diagram of this was first published in 1976 by the present writer, when its real significance was not appreciated. During this same period, outside cylinder 4-4-0s were proposed using existing Pickersgill standard boilers and cylinders, and a further development outlined in January 1918 was to have outside Walschaerts valve gear, which was subsequently drawn out in some detail.
E.S. Cox, whose enquiry had provoked the first revelations in 1948, when he himself was planning a range of standard locomotives for the newly-formed British Railways, later expressed doubt that the LYR, with which he had commenced his railway career in 1917, had made any positive proposals to the ARLE. Unfortunately the Horwich drawing registers no longer appear to exist, but diagrams must have been produced, because these were photographed on to still-surviving glass plate negatives on 9 October and 15 November, 1917. These were for outside cylinder 4-4-0, 2-6-0, 4-6-0 and 2-8-0 locomotives, a curious feature of which would have been that all would have carried precisely the same design of boiler. Having a diameter of no less than 6 ft and extending 13 ft between tubeplates, with a round-top firebox affording 32 sq ft of grate area an estimated 18 ton axleload for the 2-6-0 was somewhat optimistic! Copies ofthe Horwich schemes were logged in the Swindon drawing register in December 1917. Similarly, copies of the Doncaster and Derby 2-8-0 schemes had been logged on 21 August, only one week after their preparation, but there was no record of receipt of the Gorton and St Rollox 2-6-0 diagrams.
The proposals which came closest to actual adoption were those from Ashford and Derby. There was something of a common denominator here in that the Chief Draughtsman at Ashford was James Clayton, who according to Holcroft was the kingpin of the whole locomotive standardisation project. Before arriving at Ashford in late 1913, recruited by the newly appointed R.E.L. Maunsell, Clayton had been a leading draughtsman with the Midland Railway at Derby, where he had initiated the design of the celebrated Somerset & Dorset 2-8-0. A development of this with longer boiler was sketched out in August 1917 at Derby, which made no corresponding proposal for a 2-6-0.
For its part, Ashford had only recently produced the prototype Maunsell 2-6-0, No. 810, which could justifiably claim to be the most advanced mixed traffic engine in the country, with its taper boiler, high superheat, and long travel piston valves.
In November 1917, Ashford DO offered several possible variations on the 2-6-0 theme, ranging in weight between 61 and 65 tons. Variables were taper boiler or parallel boiler, with 5 ft 6 in or 5 ft 8 in coupled wheels. That selected was the combination of parallel boiler with 5 ft 8 in wheels. (It will be noted 5 ft 8 in was a Swindon and Doncaster standard, which had also featured on the CR, GCR, and LYR 2-6-0 proposals.) In similar mould, a particularly handsome heavy goods 2-8-0 was worked up in January 1918, which would have had many components, including the cylinders, in common, which one can only regret was never built.
It would have been reasonable to assume that these two related designs would both have been approved for production, but no. The somewhat austere Derby 2-8-0 was provisionally selected instead. A pencilled note on a surviving diagram records that this was to be presented at a meeting to be held on 22 November 1917, when its wheelbase (engine only) was shown as 8 ft 6 in + 6 ft + 5 ft 6 in + 6 ft. This was subsequently amended to 8 ft 3 in + 5 ft 6in + 5 ft 6 in + 6 ft 6 in and was approved on 12 February 1918.
One presumes that at the latter meeting there was some debate as to whether the new engines should have short lap/short valve travel as still adhered to at Derby, or long lap/long travel as favoured by Ashford. Eight days later, Ashford produced comparative valve diagrams for short (43f16 in) and long (67/16 in) travel valves. The question was evidently settled one way or the other by late June 1918, when Churchward declared that "the main parts, such as cylinders, piston valves, motion, valve gear etc. will be interchangeable, and the weight will not exceed that which can be accepted by certain of the Companies:".
On the subject of weight 17 different railways, which ranged in size from the Cambrian to the LNWR, had submitted by mid-April 1918 details of the axleloads they were prepared to accept. For the 2-6-0 the GNR, NER, and LSWR could take 20 tons, as against only 15 tons by the Cambrian. Concerning total weight the NER could take a 2-6-0 weighing no less than 75 tons compared to only 52 tons on the Cambrian.
As regards the 2-8-0, this was rejected by both the SE&CR, and Highland. (The latter had been prepared to accept an 18 ton axleload on the 2-6-0, despite having rejected the 'River' class 4-6-0s with 17¾ tons only three years earlier.) However the Great North of Scotland offered two weight options, and the Cambrian was prepared to accept a 2-8-0 provided its axleload did not exceed 14 tons!
Following the Armistice in November 1918 genuine interest in these projects rapidly faded: differences such as the loading gauges of the different companies proved in practice almost impossible to reconcile. E.S. Cox has recalled how revived interest in the possibility of a Channel Tunnel prompted consideration of the adoption in Britain of the more generous Berne Continental loading gauge at this time. Had this indeed come about, Cox's task 30 years later would have been somewhat easier when he found himself formulating plans in 1948 for a standard range of locomotives for the newly- nationalised British Railways — a national locomotive fleet at last! These were indeed produced, but were destined to have but a brief existence: all 999 were retired within all but 20 years of their conception owing to a fundamental change in motive power policy only six years later.

1. C. S. Cocks, 'History of Southern Railway Locomotives to 1938', Journal of the Institution of Locomotive Engineers, No. 206, pp 859-860, 1948. (Paper 481)
2. H. Holcroft, Locomotive Adventure, Vol 1, pp 87-89, Ian Allan Ltd., 1962.
3. E. B. Trotter. 'Standards that were never built'. Trains Illustrated February 1954, pp 60-62.
4. Journal of the Stephenson Locomotive Society, February 1951 pp 44-45.
5. C. P. Atkins, The Scottish 4-6-0 Classes, p. 110, Ian Allan Ltd. 1976.
6. E. S. Cox, British Railways Standard Steam Locomotives, pp 18-19, Ian Allan Ltd. 1966.
7. C. P. Atkins, 'The James Clayton Influence,' Railways South East, winter 1988/9, pp 122·129.
8. H. A. V. Bulleid, Master Builders of Steam p. 129, Ian Allan Ltd. 1963.
9. E. S. Cox, Chronicles of Steam, p. 12, Ian Allan Ltd. 1967.

Particulars of proposed mixed traffic 2-6-0s by ARLE members, 1917

Design Office





St. Rollox


Drawing /Sketch No.











late 1917



Cylinders, in x in

20½ x 28

20½ x 28

20 x 30

20½ x 28

21 x 28

20½ x 28

Driving wheel dia. ft-in







Boiler pressure psi







Boiler dia, ft-in







Evaporative HS, ft2







Superheater, ft2






Grate area, ft2







Adhesive wt, tons






Not given

Engine wt, tons







Rigid wheelbase, ft-in







Engine wheelbase, ft-in







Tractive force, Ibf