George Jackson Churchward

Illustration taken from Railway Magazine 1910: The Great Bear.

Steamindex Home page

Born Stoke Gabriel, Devon in 1857. Killed Swindon December 1933. His greatness was widely recognized, both by Grealey and by Stanier: the latter presented a paper about him to the Newcomen Society.
Diamond's comments based partly by conversations with Stanier are particularly interesting and may have been missed.


Own papers

Testing plant on the Great Western Railway at Swindon. Proc. Instn mech. Engrs., 1904, 67, 937-9.
Bed of cast iron bolted onto a concrete platform. Five pairs of bearings enabled horizontal travel. Dynamometer. At same meeting there were contributions from Pettigrew and from McIntosh.

Large locomotive boilers. Proc. Instn. mech. Engrs, 1906, 70, 165-256.
"In America the great power of engines now employed renders the wide fire-box a necessity, but in Great Britain, where the coal burnt per mile is very much less, few boilers of this kind have been built." He the made reference to the Ivatt Atlantics and Holden's 0-10-0T. Continuing [in the USA] " poorer coals in large quantitities can be burnt with much greater facility and economy in this type than in the narrow pattern [but his offset when] "goods trains are kept standing, as is often the case." With reference to the USA, Churchward noted the leaking of tubes and via stays. A higher standard of skill was required by the fireman. The main mass of the fire being so much nearer the tube-plate had a bad effect upon the tubes. Churchward referred to Drummond's experiments with water tubes and noted his experiments on steam drying. Churchward had fitted a Schmidt superheater to the No. 1 boiler.

On the question of economic production and use of steam locomotives. Ninth Int. Rly Congr. Subject V. Int. Rly Cong. Bull., 1910, 3
Statistics on superheated locomotives on British railways. Source: Rutherford

ARLE 1902
At the November 1902 meeting of ARLE (as edited Hughes)
Worsdell noted that the Americans had gone in strongly for compounds, but were now giving them up (this was in response to enthusiasm for developments in France as expressed by Johnson) and by Churchward who noted that a French compound had been ordered by the GWR for comparison with his simple engines..

Discussion on other's papers

Cowan, P.J.
American locomotive practice. Proc. Instn Civ. Engrs, 1903.
Pp. 400-04. According to Rutherford (Backtrack 11 622 (626): refering to the 2-6-2 constructed for the LS&MSRR stated: "belongs to a class of engine, which he feared, at no distant date would be arrived at in England: he did not see any alternative, with the train services which some companies had to run"

Sauvage, Edouard
Compound locomotives in France. Proc. Instn mech. Engrs., 1904, 66, 327-80. Disc.: 380-467.
Pp. 398-400: Disc.: 380-467.
Churchward contributed to the discussion on pp 400-4 (given in full).

Fowler, Henry
Superheating steam in locomotives. Min. Proc. Instn civ. Engrs, 196, 129-32
Churchward objected to the inferences being amde that Swindon only used a low degree of superheating. It is noteworthy that Gresley argued (139-41) in support of those who had claimed that Swindon's was only a low degree of superheating.

Address on retirement of William Dean (quoted via Armin)

'Those of them who had been more particularly employed at Swindon, and especially the older men amongst them, remembered the very great work which had been done by Mr. Dean in bringing their works to this present magnificent size and condition. Their admiration was further drawn forth when they saw the latest addition to those Works [the new A Shop]. The whole of that – practically in every detail – was designed by Mr. Dean himself." Armin states that It's difficult to be certain how much weight to place on this statement. Alan Peck in The Great Western at Swindon Works, attributes most of the work on A Shop to Churchward. However, once again it seems ridiculous for Churchward to make such a statement to a gathering of men who were intimately involved in the life and construction of the workshops them selves. We are therefore, forced to conclude that this statement is true, which, if it is so, confers upon Dean a very great accolade indeed, for the design of A Shop was very rarely equalled let alone surpassed in the age of the steam locomotive. A Shop's construction was authorised on 27rh June 1900 at a cost of £33,000 and included a new machine shop as well as an erecting shop. A subsequent decision, taken in 1901, provided that electricity should be used to generate power in the new shops. This may also have been a far-sighted decision of Dean's. He would certainly have had to authorise it himself, no matter who made the initial suggestion. It seems therefore that Dean's final two years in office were employed largely with the design and equipment of the new workshops, while Churchward worked on new locomotive developments.

The second interesting incident concerns a patent for a design of 'either side' wagon brake, which was issued on 4th December 1902 to W. Dean of Newburn, Swindon in the county of Wiltshire and another (Churchward).

3495/1894.Improvements in Steam Traps. Applied 19 February 1894. Published 19 January 1895.
22,730/1898. Improvements in engine piston valves. Applied 29 October 1898. Published 26 August 1899.
202/1902  with William Dean Improvements in either side hand brakes for railway rolling stock. Applied 3 January 1902. Published 4 December 1902
3958/1904 Improvements in automatic vacuum brakes.with Frederick William Snell, Frederick Henry Rayer; Conrad Knowles Dumas. Applied 17 February 1904. Published 26 January 1905.
Improvements in automatic vacuum brakes, for railway trains. Applied 12 October 1904. Published 31 August 1905
with Frederick William Snell, Frederick Henry Rayer; Conrad Knowles Dumas. Improvements in automatic vacuum brakes. Applied 20 October 1905. Published 30 August 1906.
Refers to 3598 of 1904.
22,177/1905:  Improvements in automatic vacuum brakes. applied 31 October 1905, published 26 July 1906.
4209/1908 Improvements in steam boiler superheaters. with George Henry Burrows and Clifford Charles Champeney. Published 4 February 1909. Applied 25 February 1908. Priority date 4 April 1907.
27,181/1908 Improvements in flue type superheaters. with George Henry Burrows and Clifford Charles Champeney. Published 15 December 1909. Applied 15 December 1908.

Three further patents which were abandoned have been traced:
14,353/1909. Lubricating steam engines. 19 June 1909
10,117/1914. Blastpipes. 24 April 1914
14,308/1914. 15 June 1914

Churchward's stature as a locomotive engineer is perhaps most readily gauged by reference to the appreciations written or spoken by his fellow engineers, especially those from Stanier (and from the number of his, or his successors', designs which have been preserved, although he would have probably disagreed with this action). His major contributions to locomotive engineering were in front-end design, especially in the introduction of long-travel valves, boiler design (the taper-boiler) and the rigorous policy of standardization. On the GWR there was hardly any departure from Churchward's policy until after nationalization and most of the motive power used on Western Region lines until the end of steam could trace its ancestry back to Churchward s basic norms—even to the extent of low degree superheating.and the strange retention of inside valve gear on large locomotives.

Stanier took Churchward's ideas to the LMS, and applied them in a virtually undi1uted form. Experience showed that they were, by 1933, in need of revision and Stanier and his successors on the LMS and British Railways developed Churchward's work from where he had left it. Other engineers, especially Gresley, Maunsell and Bulleid adopted some of his ideas, but adapted them to their own design philosophy.

Thus, Churchward's influence is a key to the whole period following 1923. Stanier's paper presented to the Newcomen Society, and K.J. Cook's paper given to the Locomotive Engineers are the two most detailed studies and are both written with the advantage of personal knowledge of Churchward. This approach through inside knowledge is also clearly advantageous in the writings of H.A.V. Bulleid (Master Builders: Chapter 4 contains much personal anectodal material, and is especially important for recording Churchward's involvement at ARLE meetings) and H. Holcroft. In conclusion, it should be repeated that Stanier's deep respect for his former master and tutor is Churchward's greatest tribute. [Jones]


The literature about Churchward is voluminous: nevertheless, Dow British steam horses produced a succinct summary of Churchward's achievements. Churchward was a year younger than Robinson, having been born in Stoke Gabriel on 31 January 1857 (Marshall). He started his railway career on the old South Devon Railway and at the age of 40 was appointed Chief Assistant to William Dean, Locomotive, Carriage, and Wagon Super endent of the Great Western, under whom Robinson had served as an apprentice. He succeeded Dean in 1902, was redesignated Chief Mechanical Engineer fourteen years later, and retired in 1921.

His first design, which appeared in 1902, was an outside cylindered 4-6-0 which possessed the unusually long stroke of 30 in. associated with the quite exceptional maximum valve travel of 6¾in. In the following year, attracted by the success being obtained by four-cylinder compound Atlantics on the Northern Railway of France, he persuaded the Great Western directors to import three of them so that their performance could be studied at close quarters; the first arrived that year and two others, slightly larger, in 1905.

The French engines were matched against the 4-6-0s and to make a fair comparison, one of the latter, Albion, was temporarily converted to the Atlantic wheel arrangement. The tests did not prove that the four-cylinder French compounds were superior to the two-cylinder Great Western simples and Churchward's next step was to build a four-cylinder 'Atlantic', which later bore the famous name North Star, fore-runner of the celebrated Star class 4-6-a's. Taking advantage of the experience gained with the French engines, which had the boiler pressure of 227 lb. per square inch, he gave his new design a pressure of 225 lb., a feature of the experimental Albion and hitherto unprecedented in this country.

North Star showed up well against the French engines but the better adhesion obtainable with six-coupled wheels led to her conversion to a 4-6-0 in 1909 and all four-cylinder locomotives since built by the Great Western, with the solitary exception of The Great Bear, have been provided with this wheel arrangement. The Great Bear was built in 1908. She was the first Pacific to appear on British metals and the only Pacific ever possessed by the Great Western, from which it can be assumed that she was not altogether a success. Indeed, all the available evidence goes to show that her construction was called for by the directors simply to give the Great Western the distinction of building the first British Pacific, as well as the biggest and heaviest engine of the day. Some Great Western men assert, however, that The Great Bear was a logical development by Churchward of his standard engine series, and that at the time it was built the object was to provide a four-cylinder machine with 15-in. diameter cylinders which it was then thought could not adequately be fed by a standard boiler. Nevertheless, owing to engineering considerations The Great Bear spent the whole of its life as a Pacific on the main line between London and Bristol, but its somewhat restricted activities were widened in 1924 when, needing a new boiler, it was rebuilt as a 4-6-0.

These early experiments of Churchward all represented integral parts of a scientific scheme of locomotive standardization. They were accompanied by the introduction of the high-pressure tapered or coned domeless boiler, the early adoption of superheating, As a result of the initiation of these advanced principles of design – and it was in their successful combination that Church yard's genius was demonstrated—the foundations of future Swindon locomotive practice for years to come were well and truly laid by about 1906. It was not until the middle 1920's that supremacy of Great Western locomotives for efficiency and speed was seriously challenged by another railway.

Rutherford (Great Western 4-6-0s p.18) notes a rumour that Karl Göldsdorf may have visited Swindon

Always ready to experiment and to learn from others, Churchward left an indelible mark on British locomotive practice, and the fact that so many of the features he introduced on Great Western were later adopted by the other British railways is ample tribute to his genius. Preoccupied with locomotives to the end, he often pottered about Swindon during his retirement, and it was on one of these visits, in December 1933 that he was knocked by one of his own engines and killed.

Westwood stated that Churchward was one of the most influential designers of the twentieth century, Churchward created a range of very advanced standard locomotives during his time as locomotive superintendent of Britain's GWR (1902-21). His masterful combination of American, continental, and Great Western features was more responsible for his success than any great inventiveness of his own. His use of long-travel valves opened up new possibilities for the more economical use of steam, although British designers responded very slowly to this lesson (and Americans hardly at all). He died on 19 December 1933 after being hit by a train.

His grave is at St. Mary & St. Gabriel churchyard in Stoke Gabriel in Devon see Humm J. Rly Canal Hist. Soc.,, 2015, 38, 252..

Carpenter, George W. entry in Oxford Dictionary of National Biography
Durrant, A.E. Swindon apprentice. Chapter 9: The Churchward legend
Adds considerably to understanding Chaurchward's international standing
Rogers, G. J. Churchward;
Tuplin, Great Western Steam;
Holcroft, Outline of Great Western Locomotive Practice.
Williams: Stars of Steam .

Nock, O.S. Railway enthusuast's encyclopedia

Arman, Brian. The Dean-Churchward transition. Br Rly J., 1994 (36) 267-79.
Summers, L.A.: G.J. Churchward and his contemporaries. Backtrack
Churchward's classification scheme. (Railway reflections). Michael Rutherford. Backtrack, 12, 50
Stanier, W.A. George Jackson Churchward, Chief Mechanical Engineer, Great Western Railway. Trans. Newcomen Soc., 1960, 30,1-8. Disc.: 8-12 + 4 plates. 4 illus. (mci. port.), 3 diagrs. See below for extract.

Cook, K.J. The late G.J. Churchward's locomotive development on the Great Western Railway. J. Instn Loco. Engrs, 1950, 40, 131-71. Disc.: 171-210. + folding plate. 33 illus., 20 diagrs., 4 tables. (Paper No.492).
This is the most complete professional source of assembled data on Churchward design. In the discussion W.A. Stanier presented some anecdotal material (pp. 171-2) whilst H. Holcroft (pp. 173-82) added a considerable amount of extra detail.

Holcroft, H. An outline of Great Western locomotive practice, 1837-1947. 1957. p. 97:

Having heard all sides of the matter discussed, Churchward might adjourn the meeting for a day or so for further development or for alterations to be made, or for an alternative scheme to be produced. When satisfied, he would sum up all the pros and cons in an entirely logical way and arrive at a clear-cut decision. One always felt that finality had been reached and the best possible solution had been found. Such was the way in which Swindon design proceeded step by step, and it put the Great Western a decade or more ahead of any other railway in the country as far as locomotives were concerned.

Churchward was a fine figure of a man and his even temperament and dignified bearing suggested the country squire, an impression which his fondness for tweed suitings heightened. He was above all a tactful admini strator and a leader of men. He gathered about him by careful selection a technical staff of diversified talents, many of them with academic distinctions~ These he inspired with enthusiasm and drew from them their best, and he created the conditions for the growth of good team work and esprit de corps.

The eminence attained by Churchward was due to his personality as much as anything. His depth of vision, wonderful grasp of essentials, logical thinking and the intuition by which he seemed to sense what was at the back of the minds of those about him, ensured that the best was always forthcoming. Such items as the taper boiler barrel, Belpaire firebox, long- travel valves, top feed and the firetube superheater had first appeared else where, while the domeless boiler with internal steam collecting pipe and regulator in smokebox was a revival of broad-gauge practice. It was by the recognition of the good points in these details and the careful blending of them into his own designs that success was attained, rather than by the originality of ideas.

He was Mayor of Swindon in 1900.

Ell, S.O. discussion on Tuplin, W.A. Some questions about the steam locomotive. J. Instn Loco. Engrs., 1953, 43, 671-4. (Paper No. 528).
Ell in his demolition of Tuplin's paper noted that "Goss does not connect blast pipe and chimney dimensions at all, which is surprising since there are practical reasons why they should be connected and no scientific reasons why they should not be. Goss ties the chimney choke to the roof of the smoke box and the orifice top to the centre line of the smokebox. These are too restrictive for British practice. In his illustration of Goss's proportions, Fig. 8, the Author omits to define the proportions of the chimney above the choke. Are we to infer that he considers this has no other purpose than to lead the gas-steam mixture to the outside? Mr. Ell would like to show the Author how the action here and in the choke is studied in modern practice when developing front-end proportions. Churchward's proportions as between blast pipe and chimney, dating from the same period as Goss's work, have never been seriously upset and it is interesting to note that Young produced in America long after Churchward and Goss, proportions which are almost identical with those of Churchward".

Gresley discussion on Collins, G.H.H. J. Instn Loco Engrs., 1934, 24, 772-4. The manufacture and repair of locomotive boiler tubes. 748-72. Disc.: 772-84. (Paper No. 330)
Gresley (772-4) referred to the stretching of ordinary boiler tubes, This practice has been introduced by the Great Western Railway and has been a very good one indeed. I myself have copied it, as I have copied many Great Western Railway practices. I remember an occasion when I went down to see the works at Swindon and got Mr. Churchward to lend me the drawings so that I could make a stretching bench for myself. Now, however, I consider that the stretching method is not the best one; the cutting off of the ends, and the flash welding of a new end is the cheapest and most effective practice to-day

George Jackson Churchward, Chief Mechanical Engineer, Great Western Railway

SIR WILLIAM STANIER, D.Sc., F.R.S., Hon.M.I.Mech.E., M.I .Loco .E., Member
(Read at the Science Museum, London, 12 October 1955)

George Jackson Churchward was born at Stoke Gabriel on the River Dart, Devonshire in 1857. His father was the Squire of Stoke Gabriel so that as a boy he enjoyed plenty of opportu nities for fishing and shooting, the hobbies which stood him in good stead in his busy mature life. He early showed an interest in mechanical devices and enjoyed making things. In 1873 he went to Newton Abbot as a pupil to John Wright, the Locomotive, Carriage. and Wagon Superinteadent of the South Devon, Cornwall and West Cornwall Railways. The main workshops of these Railways was at Newton Abbot. At this time Richard Neville Grenville was also a pupil at Newton Abbot and between them they designed and built one of the first motor cars. It was a three-wheeler and the single wheel in front was tiller steered. They designed and built a steam engine to drive it and used a small Merryweather fire engine boiler to provide steam. This was before 1876 and I believe it still exists in the possession of Allans of Oxford.

In 1876 the South Devon and Cornwall Railways were absorbed by the Great Western Railway and Churchward went to Swindon to serve the last years of his pupilage and then to the drawing office in 1877. At this time George Westinghouse came from America to introduce his air pressure brake to the English railways and it is rumoured that he visited Swindon to see Joseph Armstrong to try and get him to adopt it on the Great Western. After telling Mr. Armstrozlg what his brake would do and finding him not very responsive Westinghouse wound up by saying, "Well, Mr. Armstrong, you will have to have my brake." Mr. Armstrong replied that he would see about it and left it at that, but later he called in young Joe Armstrong, his son, and George Jackson Churchward and gave them the job of producing a vacuum brake. The Great Western vacuum brake was the result. It still uses 25 in. of vacuum and not 20 in. as all the other vacuum brakes use and the Great Western brake still has a vacuum pump driven from one of the crossheads of the engine instead of by a small ejector to maintain the vacuum when running.

Churchward was also engaged on the design of plant for the Severn Tunnel Pumping Station at Sudbrook. No doubt when he was on the South Devon and Cornwall Railways he had seen and was familiar with many Cornish pumping engines which probably had something to do with the installation of the very fine examples built by Harvey's of Hayle for the Severn Tunnel plant.

In June 1882 Churchward became an Inspecting Engineer for materials, but in December 1882 he was appointed Assistant Carriage Works Manager. At that time James Holden (who afterwards went to the Great Eastern) was Assistant to William Dean, who had succeeded Joseph Armstrong, and Dean looked on James Holden as his Chief Carriage and Wagon man. In 1885 Churchward was appointed Carriage Works Manager and it was whilst occupying this position that he developed the O.K. axlebox for carriages. It is reported that, while he was carrying our experiments with this box, he was amazed at the reports his inspectors made as to the cause of hot boxes; they always blamed dirt on the bearings. Churchward took a Carr brake van out of service, drilled a hole through the floor and the cower of the axlebox, inserted a copper pipe with a funnel at the top and then arranged for it to work in a train from Swindon to London. During the journey he fed flour emery down the funnel and on arrival at Paddington told his inspectors to feel the box. It was quite cool, but of course the journal was more like an hour-glass than a journal. He had studied the theory of lubrication of Osborne Reynolds and in the design of the box he not only saw that oil could get between the surfaces of the journal and bearing, but that there was a good pad below the journal to feed and maintain the oil film. He also designed the box so that there was proper metallic contact between the brass and the box to conduct the heat away; for he knew that provided the work was done properly, dirt should not heat the box although it might cut the journal.

In 1895 Churchward went to the locomotive works as Assistant Works Manager, and on the retirement of Samuel Carlton in 1895, became Locomotive Works Manager. He at once started modernising the machine shops by installing British and American machines more able to take advantage of the new tool steels that were gradually being introduced. Churchward had a flair for selecting good engineering things.

Up to this time the express passenger engines had a single pair of large diameter driving wheels, and up to the conversion of the gauge in 1892 the Lord of the Isles class designed by Daniel Gooch still held pride of place. On the conversion of the gauge, Dean had built and was building the very handsome 7 ft. 8 in. single-wheel engine which for many years hauled practically all the main line fast passenger trains. In 1895 he introduced the four-wheel-coupled engine for the South Devon and Cornwall work, and this class of 4-4-0 engine, known as the Duke class, was the forerunner of a series of engines all of which had the same design of cylinders, valve gear, axle boxes, rods etc. The Duke had driving wheels 5 ft. 8 in. diam., the Badminton for faster passenger trains had wheels 6 ft. 8in. diam. and the Aberdares for heavy goods trains, 4 ft. 7 in. diam. These engines undoubtedly set the pattern for the first group of standard engines using the standard boilers built by Churchward.

In 1902 Churchward succeeded Dean as Locomotive, Carriage and Wagon Superintendent; the title was altered to Chief Mechanical Engineer in 1916. The first thing that occupied Churchward's attention was the design of the valve gear for future engines, and he had a small stationary steam engine built embodying his ideas of what the valve distribution should be. In this engine he incorporated a valve gear which gave a much longer stroke to the valve and greater steam lap than was usual. He satisfied himself that he was able thus to get a much fuller opening to exhaust when the engine was notched up.

Churchward undoubtedly had a gift for selecting young men who were not only able, but were willing to carry out his ideas in actual practice. He selected Collett from the drawing office and made him Assistant Works Manager; he put with him G.H. Pearson and J.W. Cross as junior Assistants to carry out experimental and development work on the manufacture of the hardware; he had George Burrows in the drawing office to work out the designs. A standard valve gear for outside-cylinder engines was developed embodying Stephenson link motion to give a valve travel of 6¼ in. with a steam lap of l¾ in. With this gear he used a piston valve 10 in. diam., which was first used on engine No. 98. and was so successful that it became the standard arrangement of cylinders and valve gear for the whole of his series of two-cylinder engines.

{not reproduced}

Churchward was a great believer in large bearing surfaces and right from the beginning his standard engines had large journals and the bearings were so designed that they had a running clearance on - each side to build up the oil fl]m with a proper wool and horsehair pad to keep the journal flooded with oil, so that his engines were remarkably free from trouble from hot boxes.

When he retired in December 1921 the men in the works wished to make him a presentation; on being approached he told them he did not want anything, but on being pressed to accept something he asked for a fishing rod. Practically the whole of the staff in the works contributed to the presenta tion, and he was presented with a fishing rod at a mass meeting in the Mechanics Institute. On the platform were the officers of the various sections under his control and the Works Committee. The Chairman of the Works Committee gave a very eulogistic account of how the men appreciated him as a Chief and wound up by hoping that every hair on his head would be a candle to light him to glory. The old man, who was as bald as a coot, turned round and said "There won't be many of them, Watkins." It will be appreciated that, of course, the amount collected was very much more than would buy a fishing rod, and he expressed a wish that the surplus should form a trust fund to provide prizes for the two apprentices each year who had shown the most interest in their work, both in the shops and at the Technical School, and so far as I know this

In 1933 he was killed while inspecting the line which ran alongside his garden, because he had noticed what he thought were slack places in the sleepers. Unfortunately, in the latter years of his life he was deaf; and suffered from glycoma in his eyes, so he did not realise that one of the fast trains to Wales, running late, was approaching and he was knocked down and killed on 19 December 1933, mourned by all his men. So ended the life of the most progressive locomotive engineer of his time.

K.J. Cook, who was an apprentice during Churchward's later years notes that he "generally wore a suit of tweed or worstead of a medium colour and quite resembled a country squire". Churchward continued to live at Newburn, the house built for Joseph Armstrong, following his retirement.

Holcroft notes that: Seen in retrospect, the Churchward era was in two parts. During the first decade came the creation of the standard types and the accomplishment of the task to which their designer had applied himself, so putting the Great Western some ten years or more ahead of other British railways and earning a reputation for himself which placed Churchward in the front rank of locomotive engineers of all time. Resting on his laurels, he was content to watch his standard types in action and others continuously rolling off the Swindon Plant to swell the numbers thrcugh the next decade, which ended in his retirement.

The transition from Dean to Churchward at Swindon has been the cause of some comment. Peck states that "G.J. Churchward had been moved in 1898, from his post as Carriage Works Manager to that of Assiatnt Works Manager preumably because of the continued illness of Carlton, and he [GJC] was now appointed manager in his place from 25th March 1896. This placed him, in effect, but not in name, as senior assitant to Dean, and from this date one can begin to discern his hand clearly in the matter of locomotive design.

Diamond, E.L. Development of locomotive power at speed. Proc. Instn Mech. Engrs., 1947, 156, 404-16. Disc.: 417-43.
Reasons for Improvement in Performance at Speed. Considerations of space will not permit of a detailed discussion of the improvements in valve gear design which have contributed to these results, and in any case they have been extensively discussed elsewhere*. In his 1905 paper Dalby drew attention quite explicitly to the causes of the extreme attenuation of the diagrams which he studied. The only locomotive engineer in this country who at that time was following the same line of thought was the late G.J. Churchward, whose outstanding genius as a designer is perhaps only now being fully recognized. The author has been told by Sir William Stanier that Churchward set up a single locomotive cylinder and engine unit at Swindon with which he tried out a number of modifications to valve gears. The author has previously suggested that valuable research could be accomplished by this means without the necessity of a full-scale testing plant, and was unaware that Churchward had, in fact, satisfied himself as to the great advantage of increasing the steam lap of the valve by such means. A recent example of the value of such a method of testing is the successful development of the Franklin system of poppet valve gear on a laboratory test plant at Baltimore, U.S.A. (Alcock 1944).

Other appreciations

E.A. Langridge (196-7) noted how Churchward had introduced high boiler pressures (225 lb) and stuck with them unlike other locomotive engineers: he considered that was indicative of excellent workshop practice and the ability to maintain firebox stays and tubes. He queried the cylinder layout of the four-cylinder locomotives: on other railways this layout had been tried but dropped due to the loosening of the cylinders.

George Jackson Churchward. D.O. Coakham  Br, Rly. J., 1993 (47), 352
See Brian Arman's article on the Dean/Churchward partnership (BRJ 36) there seems to have been nothing but the odd anecdote to reveal the Churchward personality. G.J.C. is mentioned both as a religious man and a considerable mathematician. Does this dispose of one legend?
On page 111 (!) of his Pictorial Record of GW Architecture (OPC 1977), Adrian Vaughan illustrates The Great Bear at Paddington. The caption relates a yarn about Churchward's interest in astrology [sic] leading him to give the 'Bear' a number which 'ancient peoples believed was associated with the Evil Eye'. This was supposed to be a means of getting his own back on the GWR directorate who allegedly insisted on him building a 4-6-2.
It looks as if Churchward's mild relaxation was the study of numerology or gematria. From what one can gather, the significance of the number 111 is in being a multiple of the 'powerful' Solar number 37, along with 666 which, in spite of its unfortunate association with the beast of Revelations, had great potency among the Gnostic Christians. A jocular remark by 'the Chief' might have been seized on and misinterpreted as it passed down the ranks.

Hall, Stanley. Railway milestones and millstones: triumphs and disasters in British railway history. 2006.
Milestone: perhaps Hall over-emphasises Churchward's significance by placing him at the beginning of a chapter entitled "Mechanical engineering, 1825 to 1913": without Stephenson and Gooch would there have been a Churchward?
Tuplin, W. Great Western steam
After about the year 1890, it was merely a matter of arithmetic for Churchward to see that, unless he slipped up somewhere, he could expect to be in charge of the Great Western Locomotive Department from 1902 (Dean's retiring date) to 1921 (his own retiring date). He planned to make those nineteen years as restful as possible. No position of such responsibility can ever be really restful because of the normal jungle-warfare of industrial life. In any large organization of any type, departmental heads keep a wary eye on each other in the incessant battle for more power and this may be nearly a full-time occupation. If a big job is to be worth having at all, the technical necessities must be so completely covered that they are not a worry. The manager gets staff to do the internal work reliably without much supervision by persuading them that what is good for him is good for them and he can then concentrate on coping with the enemies from without.

Churchward recognized that the Great Western locomotive stock in the 1890s was a collection of oddities – picturesque perhaps but still oddities – and that something better in every way was going to be needed between 1902 and 1921. At the end of that period indeed, something quite different from the steam locomotive might be required. Even before 1900, electric locomotives were running on railways, the petrol engine was used in road vehicles, and the gas turbine was being discussed. What might another twenty years bring?

Churchward must have considered questions of this sort but guessed that if anything was going to supersede the steam locomotive as a haulage agent on railways, it would probably take at least twenty years to develop it into entirely practical form. What he was sure about was that he was not going to do the developing. His job was to give the Great Western reliable locomotives with the least risk or trouble for himself and that demanded that he should not use any device or design that had not already been thoroughly tested by somebody or other. He was wedded to the very sound engineering principle of using other people's experience and he declined to lose time, money or sleep on trying to make sense out of revolutionary ideas however bright. He kept an inquiring eye on locomotive practice all over the world with the idea of combining the best from everywhere as a basis for the next generation of Swindon locomotives. The ideal for him was to have done all his experimental work while Dean was there to take the responsibility and, soon after his own acceptance of the head position, to bring out new standard designs that he could claim as his own, to cover Great Western needs until 1921. What happened after that was for someone else to worry about.

Churchward's appointment as Locomotive Superintendent in 1902 thus marks the great turning point in Great Western history. He quickly developed at Swindon standard locomotives markedly different in appearance from any others in Great Britain and of an excellence that was tardily equalled but never surpassed. He may perhaps be called the greatest locomotive engineer since Stephenson, but even that does not make him into the supernatural mechanical genius that some of his admirers are ready to assume him to be. The truth is that Churchward locomotives include no important feature that had not been previously used by someone or other, for what Churchward did was to sort out from well-tried practice the features that would best suit Great Western needs and then to make them work.

Something fairly drastic had to be done to unify the Great Western locomotive stock for it is no serious exaggeration to say that there were hardly half-a-dozen engines exactly alike. For example, it is said that there were twelve different designs of bogie in the eighty Dean "singles" which were of a standard type, as the G.W.R. had understood standardization at that time. Quite clearly a more rigorous policy was required. . . .

The advantages of reducing the locomotive stock to the smallest possible number of classes and of making components interchangeable between different classes are so great that such standardization had clearly to be the main characteristic of new locomotive practicef or the Great Western Railway. But a stock of some 3,000 odd locomotives cannot be replaced overnight and Churchward realized that some fifteen or twenty years might elapse before the new standard engines could form the bulk of the company's stock. Obviously the new engines had to be designed to do what would be required twenty years ahead, so far as that could be foreseen.

In designing the future standard engines, Churchward compromised between rash experiment and perpetuation of old methods. He accepted the broad principle of the steam locomotive but proceeded to design it with more open-minded care than had ever been used before. Far from indulging in entirely novel ideas he tried nothing that had not already had some success on the G.W.R. or elsewhere, although in size, shape and detail his first locomotives were as different from their Swindon predecessors as could be imagined.

The results achieved in the experimental engines were so gratifying that the principles embodied in them were adopted with little modification in standard designs that even over 25 years later were superior to their British contemporaries. This, of course, has often been pointed out, and it is natural to assume that the admirable qualities of Churchward locomotives are attributable entirely to their design. In actual fact, the credit must be divided between

  1. excellence of basic design;

  2. excellence of detail design;

  3. excellence of materials used;

  4. excellence of workmanship;

  5. intelligence and enterprise in handling on the road;

  6. high standard of maintenance in service.

It would obviously have been foolish to allow the overall performance of the locomotives to be vitiated by imperfection in any one of these respects and as Churchward had control of all of them he was able to ensure that every department concerned kept up to the mark. On some railways in more recent years, the organization is such that those responsible for (1) and (2) have no official control over the other matters and the results achieved usually suffer from the effects of the normal interdepartmental enmities.

Churchward started out with the big advantage that he himself could decide all major issues at every point in the locomotive's career from conception to final decomposition. With.his position assured in that respect he had no need to assume the unreasoningly . autocratic manner that some other locomotive engineers of the period are understood to have affected. In the drawing office he allotted different men to specialize on different components of the locomotive. On questions of manufacture, officials from the works were consulted while the opinions of shed superintendents and locomotive inspectors were sought on matters concerned with the daily running of the engine. He himself, with experience in every branch, could reach a sound decision on any point after hearing the views of the specialists and was usually able to convince all concerned that it was indeed sound. In the exceptional case he had the authority to enforce the decision and this eliminated the loss of time and effort that always results from reluctance to co-operate in a joint effort. It must have taken more than persuasion to get the running-shed staff to agree to the multiplication of the 4-cylinder engines with their buried valve-gear.

You will sometimes hear an engineer say pathetically "All engineering design is at the mercy of the junior draughtsman', and truly designs are to be seen that confirm that remark to the hilt. It was not so with Churchward. Every detail was examined and criticized before a design was approved and the result is seen in the elemental simplicity of the major and minor components of Swindon locomotives. Nowhere does one see fussiness, flourishes or fads as any such puerility that may have raised its head in the first stage of a design was ruthlessly swept away. It was first decided just what each component had to do, it was then designed to do just that at minimum total cost in material, manufacture and maintenance. If it looked crude or inelegant to some Victorian designer, that was unfortunate but it was not altered on that account.

Whilst the practice of discussing design in collaboration with representatives of all departments concerned is obviously advantageous, it must be admitted that it does not always have the maximum possible effect. One reason is that no single representative of a department is certain to know every detail of what occurs in it. For example, two long-established details of Swindon practice – smokebox-regulator and screwed-in piston-rods – were adopted by Stanier on the L.M.S. but were abandoned there because they could not be made to work satisfactorily. Evidently some uncoimmunicated detail of "know-how" made all the difference between success and dissatisfaction.

Another reason is that the enginemen learn to cope with a difficulty till they cease to be worried by it and after a time no complaint or suggestion for improvement in the matter ever reaches the Chief Mechanical Engineer. An example is the continued laborious use of the flap-plate as a fire-door on Great Western locomotives, when something far less fatiguing is easily practicable.

Churchward early decided that fewer than a dozen classes of locomotive would cover all Great Western needs. Starting at the large and spectacular end of the range, what type of express passenger locomotive would be required for the next twenty years? Looking at what was being done in America, thinking of the steep gradients in some parts of the Great Western main lines, anticipating increase in train loads in conjunction with the higher speeds that the G.W.R. was contemplating, he selected an outside-cylinder 4-6-0 with a grate area of about 27 sq. ft. and twenty-three years later, nothing better for the same purpose had been seen in Great Britain. Ten major.dimensions for each of six projected classes of locomotives were laid down in a document prepared in the year. 1901 to define the design programme for the new standard engines, and except for the development of 4-cylinder engines for the fastest trains, and the addition of three other classes incorporating standard major components, remarkably little departure has ever been made from this scheme in Great Western locomotive practice.

The immediate aim was to produce a few designs of boilers, cylinders, connecting rods, valve gear and wheels which all the standard engines would eventually use and to try them out in experimental engines. Actual service would suggest ways of improvement (it always does) and when satisfactory results were achieved, the new designs would be accepted as standards and very strong resistance shown to any further suggestion for modification. It is here that great willpower is required by an engineer intelligent enough to see that improvement is always possible. Having reached a satisfactory design he must say to himself, in the interest of the great economies effected by standardization, "We will make no change for at least ten years." The distinction of Churchward's work on the GWR was that in a period of about three years, intensive effort produced designs far in advance of those to be found elsewhere, and of a nature that permitted longterm standardization to be started with confidence. . . .

Not many personal details have been published about Churchward but it can be said that whilst he was held in high regard by his assistants, he was not the supernatural deity that some fervent admirers of his locomotives are apt to assume him to be. A photograph reproduced in the Railway Magazine at the time of his retirement in 1922 shows a dignified scholarly veteran whilst one that appeared in the Locomotive Magazine for January 14, 1922, depicts him as a sporty gent with a startling black moustache. Stories from those who worked for him tend to support the suggestion that he was a multiple personality. He could choose men for jobs and get the best out of them; he was in fact a leader. Churchward bore himself with dignity when the occasion demanded it but more often his manner was hearty and he could talk to labourers in language that they easily understood. He threatened to resign when the General Manager questioned the cost of Swindon locomotives. He was known to speak slightingly of one of his assistants in the presence of juniors. In short he had a full share of human frailties and that his work made history was merely incidental to its success in earning him a living by holding down,a big job on the Great Western Railway. . . .

Churchward continued to live at Swindon after his retirement and was a frequent visitor to the works. Not in his lifetime or even for long afterwards, were any substantial departures made there from his methods and so his old age was not sorrowed by any suggestion that he was being out-moded. On the contrary, the successes of Swindon engines in locomotive-exchanges led to the adoption by other British railways of features that had come to be regarded as peculiar to Churchward and so began one of the brightest periods in 'the history of British locomotive performances.

So it was that Churchward could maintain a gratified interest in Swindon and its locomotives even when responsibility for them had passed from him and so it happened that when walking from his home to the works on the foggy morning of December 19, 1933, in the seventy-fifth year of his age, he was run down and killed by a train on the main line.

Duffy, M.C. Technomorphology and the Stephenson traction system. Trans. Newcomen Soc., 1982, 54,55-74. Disc.: 74-8.
Notes that Churchward understood the significanace of Goss's work at an early stage.
H.M. Le Fleming (Concise encyclopaedia)...

Blowing the Swindon myth
Without doubt Churchward's successor was secretive and delibertaely uncooperative with his colleagues, and that this may have been a "continuation" of Chuchward's policies. This is without foundation as Stanier's and other more junior members of Churchward's staff in the Transactions of the new Institution of Locomotive Engineers clearly shows (below).

Fry, Lawford H.
Modern locomotive practice in Europe and America. Trans Instn Loco. Engrs., 1913, 3, 41-58. Disc.: 59-73. (Paper No. 16)
Stanier (63-4) noted the existence of the Churchward 2-8-0 and 2-8-0T on the GWR. The leading coupled wheels of the 42XX type had spherical crank pins and bushes to prove 2½in side play: this enabled the class to operate over sharp curves at collieries. Stanier noted that the 4-cylinder layout distributed the working stress "very nicely". Refering to the 43XX he called the mixed reaffic locomotive the "engine of the future" noting that is was suitable for working excursion trafic and troop trains.

See also Great Western locomotives

Updated 2015-11-15

Return to Home page