Great Western 4-4-2 Locomotives in Great_Britain

Class 4600 (Locobase 20347)

Data from "4-4-2 Tank Locomotive, Great Western Railway", Locomotive Magazine, Volume XX [20] (15 May 1914), p. 120.

This small Mogul tank had the trademark coned boiler and Belpaire firebox of most GWR engines. Designed for the suburban services around Birmingham, it appears to have been the only locomotive in its class.

Class County Tank (Locobase 9176)

Data from Maurice Demoulin, Locomotive Actuelle ... (Paris: Librairie Polytechnique Ch.Beranger, 1906).

These Belpaire-boilered engines were essentially County 4-4-0s that were given tanks and a trailing truck to carry them. They had a short wheelbase that, together with large external cylinders, contributed to their reputation for rough riding.

Locobase 2311 shows the superheated version that proved unsuccessful.

Class County Tank - superheated (Locobase 2311)

Data from "4-4-2 Tank Locomotive, Great Western Railway", Locomotive Magazine, Volume XX [20] (15 May 1914), p. 120. See also [] (first accessed in 2000) for details on this class, including all names.

Locobase 9177 shows the original, saturated-boiler locomotive that was built between 1905 and 1912. This class was later superheated (retaining their Belpaire firebox), but the minute amount of superheat surface leads one to wonder why they bothered with the complexity. These tanks proved too small for the work they needed to do and were scrapped in the early 1930s. Daniel (2000) adds that they rode roughly (similar to the County 4-4-0 tender engines) because of the large reciprocating masses. A short wheelbase and a long cylinder stroke are also cited as reasons.

Class La France (Locobase 10413)

Data from "New Compound Locomotive, Great Western Railway," The Locomotive Magazine, Volume IX (28 November 1903), p. 307; "Express Engines 'Albion' and 'La France'; Great Western Railway", Railway Engineer, Volume 25, No 4 (April 1904), pp. 125-127; and Charles Rous-Martens, "The French Locomotive for the Great Western Railway", Engineer, Volume 96 (4 December 2023); and "GWR 102 La France" in Wikipedia at [], last accessed on 9 December 2023.. See also M. Edouard Sauvage, "Compound Locomotives in France", excerpt from Proceedings of the meetings of the Institute of Mechanical Engineers, 18 March and 15 April 1904 (Westminster, 1904), pp. 327-467, Churchward comments pp. 400-403. (Thanks to Alexander Blessing for his 19 November 2023 commenting on driver weight led to Locobase's expanding the entry to better place the La France in British steam locomotive history._ Works number was 5409.

As part of Churchward's investigation of four-cylinder compounding, he imported three locomotives from the builder that supplied many of France's compound express locomotives. Two of them appear in Locobase 1496 and La France, which had slightly smaller cylinders and a larger boiler. Like other GWR engines, this import had a Belpaire firebox. On the other hand, the 102 used the large-diameter, internally finned Serve tubes and, noted Charles Rous-Martens, a 1 1/2 ton higher axle loading.

The author of the Wikipedia entry notes that the de Glehn system did confer a smoother ride and less wear on the valve train. More important for later development on the GWR, Churchward found the divided drive a very useful feature. According to a King Edward II Project discussion of Churchward's several innovations, although compounding yielded little advantage in consumption, "the smoother riding of the 4-cylinder engine encouraged economical use of steam by crews (who had tended to use long cut-offs to dampen rocking in two cylinder engines), and also gave huge scope for more power. The problems of loads on rods and axleboxes was also reduced by the French division of the drive across two sets of wheels. The concept of the GWR 4-cylinder engine was born."

During tests in early 1904, La France's coal consumption amounted to 37-40.4 lb/mile (10.42-11.32 kg/km). Also, de Glehn's bogie design influenced Great Western and later British Railways practice. Churchward reported on the early days of the La France, graciously acknowledging the quality of the design:

"He would like to take the opportunity, as both Mr. de Glehn and M. Sauvage were present, to say that "La France" had been doing very first-class work indeed on the Great Western Railway. She had given every satisfaction, and had entirely fulfilled his expectations; the work she had been doing on some of the fastest trains was really very fine, and he thought had not been equalled by any of what he might perhaps be permitted to term the old-fashioned simple engine. Those present who had studied the actual draw-bar records, which to his mind was the only record that was worth talking about in regard to a locomotive, knew that a steady pull on the draw-bar at the back of the tender of 2 tons at 70 miles an hour on a 6-foot 6-inch wheel, took, if he might use a colloquialism, a great deal of getting, and when they had it, it took a good deal of keeping up. But it was fair to say that the Great Western had two or three engines running today that would do that, and "La France" was one of them."

On the other hand, Churchward explained British resistance to the high boiler pressure and divided drive in some detail:

To the question of boiler pressure, Churchward "was not quite sure that that was necessary, but at any rate it was universal up to the present; and whilst they were still obliged to use such a pressure as 225 lbs. to the square inch, he believed that the progress of the compound principle would be checked very largely among locomotive engineers who feared trouble with the boilers. The Great Western were facing the trouble, and had a large number of boilers running today at 200 lbs. to the square inch; and in some cases, at any rate, there was evidence that the increase of trouble with the higher pressure was not quite so much greater than they had been led to expect. But still there it was, and it was impossible to get rid of it."

He then recorded his doubt about the smoothness of the de Glehn compound:

"The fact had been mentioned frequently that the divided engine and four cylinders gave a more equal torque on the shafts. He might be very dense, but he had been unable to realise that up to the present. If there were two cylinders exactly opposite one another at 180°, and the same steam distribution was used as with one, he hardly saw how the torque on the shaft could be otherwise than the same in each case."

Finally, Churchward rebutted the claim that compound could yield more power at high speed.

[S]ome of the advocates of compounding had perhaps laid too much stress on the fact that the same power could not be got at high speed out of a simple locomotive without using excessively large cylinders. The Great Western had already gone far enough to prove that one could get the necessary power without going beyond an 18-inch cylinder, and whilst the cylinder diameter was restricted to 18 inches, the stresses were really no greater on the big end and on the shaft than the dimensions they were able to provide would meet."

The engine itself remained in service until 1926.

Class North Star (Locobase 5383)

Data from Edward Cecil Poultney, British Express Locomotive Development (London: George Allen and Unwin, 1952), p. 70-75; "Four-Cylinder 4-4-2 Express Engine, Great Western Railway", Railway Engineer, Volume 27, No 2 (September 1906), p. 272; and Charles S Lake, "Locomotive Details--The Deeley Valve Gear", Model Engineer and Amateur Electrician, Volume 22 (10 June 1910), pp.562-563; and "No. 40 North Star: 1906" in steamindex at [], last accessed 3 December 2023.. (Thanks to Alexander Blessing for his 19 November 2023 email describing his research into the Deeley Valve Gear and using Churchward's data to correct several values.).

GJ Churchward design from the same year as the De Glehn compounds, but considerably smaller. It was preceded by the temporary modification of the Ten-wheeler Albion as an Atlantic.

Even though North Star was smaller than the de Glehn compounds, it still put more weight on the drivers than did the French design. More important, it featured four cylinders of the same diameter, two inside over the bogie's front axle and driving the front coupled axle, two outside over the rear axle of the leading bogie to turn the rear coupled axle. Each cylinder was fed by an 8" (203 mm) valve driven by Walschaert gear inside that actuated horizontal levers for the outside gear.

According to Alexander Blessing, "Deeley's design is an eccentric-less variant of Walschaerts, similar to Young ...Churchward apparently got in trouble for using this gear on North Star (patent infringement). ". MM&AE described the setup in detail, reporting "The steam distribution obtained with this gear is stated to be excellent in all positions of the gear." The absence of eccentrics inside meant a strengthened crank axle as well. "The gear, in fact, gives a valve movement, which is identical to that of Stephenson [the British gold standard of the day], except as regards the lead, and, morevover, it is without eccentrics."

Blessing also directed Locobase to steamindex, which referred to James Clayton's discussion of three-cylinder locomotives. Clayton spoke of two decidedly negative features of the Deeley design: " It was a good gear of the "Walschaert" type, but, unfortunately, differing from it by having the right-hand valve-gear different to the left-hand, and so added considerahlv to the cost without making the gear any better as such, though it does avoid the use of eccentrics. It also makes a hreakdown on one side of the engine a total disablement, as the motion on either side is dependent on the other."

The profile had an upright look, enhanced by the squared shoulders of the Belpaire firebox, domeless boiler, tapered safety valve stand, and relatively widely spaced leading bogie. Although the four-cylinder layout conferred both power and smoothness to express running, two driven axles were simply too few to avoid slipping if driven at the pace the design merited.

In 1909, this Atlantic was rebuilt as the lead locomotive in the Star-class 4-6-0s (Locobase 1497).

Class President, Alliance (Locobase 1496)

Data from A T Taylor, Modern British Locomotives (London: E & FN Spon, Ltd, 1907), p. 22; Charles R King, "New Four-Cylinder Compound Locomotives for the Great Western Railway of England", Railway Age, Volume XLI [41], No 10 (9 March 1906), pp. 350-351; See also Charles Rous-Marten, "French Compounds on the Great Western Railway", The Engineer, Volume (2 February 1906), p. 105-106; and O S Nock, Railways at the Turn of the Century, 1895-1905, p. 143. (Thanks to Alexander Blessing for his 24 December 2021 email correcting valve gear,

The latter two examples -- President and Alliance -- of a four-cylinder de Glehn compound (the engineer was a British expatriate) built by a French company and on a British railway. King commented that their design came from a pair of Atlantics built for the Nord as 2-641-2-2642 (Locobase 3599). OS Nock (RWC III, pl 102) points out that although these engines didn't spark the use of compounds on the GWR, they had two lasting effects. One was the bogie, which the Great Western adopted and which became an LMS standard more than 20 years later. The other was the use of de Glehn's design for the "big end" of the inside connecting rod.

The earlier de Glehn -- La France, 102 -- arrived in 1903. Its adhesion weight was 74,480 lb and engine weight was 144,816 lb. Its driving wheelbase was 7 ft 5/8" and total engine wheelbase 27 ft 10 5/8". Rous-Marten noted the Belpaire firebox's "great size"

Charles Rous-Marten, well-known for his locomotive-ride reports, rode the President on several runs; his detailed narrative arrived at tepid approval: "'[H]er performance cannot be characterised as an unqualified success. She did a great deal of very fine running, but necessarily the merit of her achievement was materially blemished by her excessive slipping, by her loss of speed up the Hemerdon bank, and finally by her breakdown at Bristol." It was, however, "only fair to point out that these mishaps imply no defect on the part of the engine herself."

A later trip on the Alliance's footplate was "in all respects a successful one, the work being invariably good and sometimes brilliant." Rous-Marten dismissed the claim that compounds could be sluggish, at least as du Bousquet and de Glehn compounds were concerned. Moreover, even when running eight quarter-miles at a 10-second pace, "the running was so admirably smooth and steady that no one ...would have been aware any higher speed than usual was being done."

Whatever the verdicts of Rous-Marten or GWR engine crews, the Atlantic arrangement could not put enough adhesive weight on two axles. The railway neither built nor bought any more 4-4-2s.

NB: A diagram published in the 1906 edition of the Locomotive Cyclopedia of American Practice (figs 5091-5092) shows two calculations for the boiler that reflect the use of Serve finned tubes. The outside dimensions, identified as the "heat distributing surface" included 1,444 sq ft (134.15 sq m) of tube heating surface. When the surface area of the fins that ran longitudinally along the inside of the tubes was included, it was termed the "heat absorbing surface" and that figure (2,582.72 sq ft/239.94 sq m) makes up the heating surface. One can make an argument either way.