Giants of Steam

by Jonathan Glancey

  *

  The latest ships and power stations at the time No. 10000 and Fury first steamed were worked with high-pressure boilers feeding steam turbines. Invented by Charles Algernon Parsons in 1884 from an idea that had originated with Hero of Alexandria, the multi-stage steam turbine was, when combined with vacuum condensing, one of the most efficient of all power units, and remains so. Could its smooth-spinning sorcery be made to work effectively in railway service? A small experimental 0-2-2-0 tank engine designed by Professor Giuseppe Belluzzo had been built as early as 1907, by Officine Meccaniche SA of Milan. Four turbines drove the 26 ton engine’s four wheels independently through a series of reduction gears. A conversion of a locomotive dating from 1876, the Belluzzo prototype was not a success; it was dismantled in 1921. Belluzzo himself went on to design two more turbine locomotives for the Italian railways. One was a 2-8-2 with high- and low-pressure turbines, built by Breda in 1931; the other, in 1933, was a rebuild of a class 685 2-6-2, with a single forward turbine, which made a number of runs between Florence and Pistoia. Both locomotives led shadowy lives and quickly disappeared. As for their inventor, Belluzzo – a committed fascist – he became a minister in Mussolini’s government. From 1942, he worked closely with the German SS on the development of a turbine-driven flying bomb which, had it flown, might have been mistaken for a flying saucer.

  Various other more or less complicated attempts at a practical steam-turbine locomotive had been made by the time William Stanier, by now chief mechanical engineer of the LMS, became interested in the idea. The most successful had, in fact, been the Beyer Peacock Ljungström 2,000 hp twin-unit condensing turbine of 1926, which had been tested by the LMS on 600 ton fast freight trains from London to Leeds in 1928.

  Stanier sailed to Sweden with representatives of the turbine division of the Metropolitan Vickers Company to see the three 1,270 hp 2-8-0 non-condensing steam-turbine locomotives designed by Birger and Fredrik Ljungström to work heavy iron-ore trains on the Grängesberg–Oxelösund railway. Built between 1930 and 1936, these were successful machines, kept busy at work until the line to the port at Oxelösund was electrified in 1953; all three have survived into preservation. Impressed by them, Stanier commissioned a turbine version of his new four-cylinder Princess Royal Pacific, based on the Ljungström locomotives. An elegant and well-resolved machine, 6202 did not look like an experiment and it achieved considerable success. In later years, when he was chairman of Power Jets, the British government’s gas-turbine organization, Stanier told Joseph Cliffe, a Doncaster-trained mechanical engineer on his team, that he would have liked to have had fifty turbine-driven Pacifics.

  The Turbomotive, as 6202 was known, proved to be slightly more efficient in terms of thermodynamics and power distribution than the Princess Royals. On test, she beat the four-cylinder Pacifics, if marginally, in terms of water and coal consumption, and she climbed Shap and Beattock more rapidly and with less chance of slipping at the head of heavy express trains. The Turbomotive was also smooth, powerful, and easy to drive. Nominally, the forward turbine with its eighteen rows of blades developed 2,600 hp with the engine travelling at 62 mph. The driver could select between one and six nozzles each blasting steam to the turbine and so raising or lowering the power output. Rotary turbine drive offered smooth torque characteristics, ensuring smoother and more slip-free starts and hill-climbing. A small, four-bladed reverse turbine was engaged by a clutch. Design complexity was kept to a minimum.

  Rather than the fan of the Beyer Peacock Ljungström locomotive, a conventional blast-pipe was fitted to exhaust steam and generate draught for the fire, while the vacuum condensing equipment that had been specified by Beyer Peacock was dispensed with, providing a major simplification at the expense of lower thermal efficiency. Soft exhaust from the double chimney clouded the crews’ view ahead and so smoke deflectors – the first on an LMS Pacific – were soon fitted to lift exhaust clear of the cab. The boiler, pressed to 250 psi, was similar to those fitted to the reciprocating Pacifics, but with a larger superheater. A maximum tractive effort of 40,000 lb was maintained at a slightly higher level than with the Princess Royals at comparable speeds, with the result that the Turbomotive was slightly more powerful on hills and at higher speeds than the reciprocating Pacifics.

  The Turbomotive made quite a name for itself on its regular runs with the fast and heavy 08.30 Euston to Liverpool Lime Street and the return working, the 17.25 Liverpool to London express, of the mid- to late 1930s. Often loaded to 480 tons, the London-bound train was booked to run the 152.6 miles from Crewe to Willesden at an average speed of 64.5 mph. A clear run tended to be the exception, so locomotives assigned to this important train had to be able to accelerate briskly from signals and sprint when necessary. O. S. Nock recorded an occasion when 6202 made up five minutes by running the 61.6 miles from Weedon to Wembley in 47.5 minutes, at an average speed of 77.8 mph.

  Unsurprisingly for a prototype, there were teething problems. The main forward turbine spindle broke one day when 6202 was bowling along the west coast main line near Leighton Buzzard at 60 mph; luckily, the turbine casing remained intact. Occasionally, a failure in the transmission system would cause the driving wheels to seize up and main-line services would be severely delayed until the front end could be jacked up and the locomotive towed away. There were also oil leaks from the enclosed triple-reduction gearing and transmission shafts, a detail Bulleid might have been aware of – this was to be one of the bugbears of his radical Merchant Navy and light Pacifics.

  Inevitably, standards of maintenance fell as the Second World War ground on. The Turbomotive was out of service, awaiting parts, and her annual wartime mileage was just 28,500 miles, compared with the 53,000 miles run by the Princess Royals and 73,000 miles by the more powerful Coronation Pacifics. She had, however, run 66,200 miles in 1938, a very creditable achievement for a prototype. When one stormy wartime evening, Cecil J. Allen needed to catch a heavy combined Liverpool and Manchester express from Crewe, he feared the worst when he saw the Turbomotive rolling into Crewe with the Liverpool portion of the train. ‘I’m afraid we’re going to lose time,’ he told his engineer companion. To Allen’s surprise, 6202 gained 11.5 minutes between Crewe and Watford at the head of its seventeen-coach, 610 ton train.

  In 1946 Roland Bond, then deputy chief mechanical engineer of the LMS, read a paper to the Institution of Locomotive Engineers on the railway’s experience with 6202. Despite teething problems, the Turbomotive had been much admired. ‘On present indications,’ said Bond, ‘the next five years should show up the turbine locomotive in an increasingly favourable light. It has, in the author’s opinion, already proved itself well worth persevering with, and it is not beyond the bounds of possibility that a limited number of non-condensing turbine locomotives, in a more highly developed form based on the experience with No. 6202, will be regularly employed on the heaviest and fastest express trains, with profit to their owners.’

  It was not to be. Having run 440,000 miles, 6202 was taken out of service in 1949, re-emerging from Crewe on 15 August 1952 as a conventional reciprocating four-cylinder Pacific. Numbered 46202 and named Princess Anne, this elegant engine – a happy cross between a Princess Royal and a Coronation – was destined for a tragically brief life. On 8 October 1952, she was one of four locomotives involved in a horrific multiple crash at Harrow and Wealdstone, 11.4 miles north of Euston. The late-running 20.15 Perth to Euston sleeper, hauled by Coronation Pacific 46242 City of Glasgow, passed signals set to danger at 60 mph and smashed into the rear of the 07.31 Tring to Euston stopping train, standing in the station behind a Fowler 4MT 2-6-4T; a few moments later, the double-headed, fifteen-car 08.00 Liverpool and Manchester express, climbing up from Euston at between 50 and 60 mph, rode into the wreckage. Its locomotives were the Jubilee class three-cylinder 4-6-0 45637 Windward Islands and 46202 Princess Anne. Both locomotives were written off and scrapped. The former Turbomotive had run just 11,442 miles in its new and regal guise. The accident
was the worst of its kind in English railway history: 112 people died and 340 were injured.

  The steam-turbine railway locomotive did appear to promise great things: smooth power; a solution to the problem of the ‘hammer blow’ on the tracks made by large reciprocating locomotives; lower fuel and water bills. In practice, however, most seemed to be beset by problems, some mechanical, others political. French and German locomotives built from the late 1920s to the early 1940s were caught out by the Second World War before they could be fully evaluated. Some, like the ungainly 2,000 hp Krupp-Zoelly condensing turbine locomotive T18 1001, a Pacific built in 1924, and the 2,700 hp Maffei condensing turbine engine T18 1002 of 1926, were damaged by allied bombing. Others, like the SNCF’s 232Q1 built by Schneider and Creusot in 1940, and resembling a rather dumpy Gresley A4, seemed promising. On test between Le Creusot and Montchanin, the 4-6-4 produced 2,600 hp and ran freely at up to 140 kph (87 mph). It was sabotaged beyond repair by retreating German troops on 3 September 1944. What also told against it was its fuel and water consumption, which was some 10 to 15 per cent higher than Chapelon 4-8-0s and 141Ps.

  One turbine engine, influenced by Stanier’s Turbomotive, that did look very promising was the Pennsylvania Railroad’s S2 6-8-6, built by Baldwin in 1944, with turbines by Westinghouse Electric. This massive and purposeful-looking machine, with a conventional fire-tube boiler pressed to 300 psi, 5 ft 8 in driving wheels, and a tractive effort of 70,500 lb, generated no less than 6,900 hp and powered thousand-ton trains happily at 100 mph and more on the lightly inclined 280 mile Crestline to Chicago main line. On a visit to the United States in 1947, E. S. Cox rode the footplate of 6200 on this section and a top speed of 100 mph was maintained for twelve minutes before the train had to be slowed.

  Although 6200 showed up well at high speed on the main line, it suffered from a problem shared by all these experimental turbine engines. Relatively efficient at full speed, their turbines were inefficient at low speed, while their rotary inertia was such that sudden changes in speed were both difficult to obtain without heavy use of the engine’s brakes and potentially damaging to the driving mechanism. Turbines are best run – as they normally are in power stations and ships – within a narrow speed range. Despite its undeniable power and speed, 6200 was withdrawn in 1949 and scrapped.

  *

  Other attempts to use steam turbines using electrical transmissions and traction motors proved to be fascinating, if complex. The first attempt had been Britain’s very first turbine locomotive, the Reid-Ramsey Turbine Electric, built by the North British Locomotive Company in 1910. With a conventional boiler at one end, capped with a bulbous chimney enclosing an exhaust fan, and condensing equipment at the other, this curious 4-4-0 + 0-4-4 machine had a push-me-pull-you appearance which lacked, to say the least, the gentle grace and refined elegance of contemporary express passenger locomotives. The boiler fed an impulse turbine coupled to a dynamo which energized four 275 hp traction motors. The locomotive was not a success and in 1924 it was rebuilt into the Reid-Macleod Turbine, equipped with high- and low-pressure turbines, but without the complication of electric motors, and promising 1,000 hp at 60 mph. This machine was shown at the British Empire Exhibition at Wembley in 1924 and perhaps, with its submarine-like profile, it may have had the look of a locomotive of the future. Trials made between March 1926 and April 1927, however, were decidedly disappointing. Its one big chance was a run from Glasgow to Edinburgh and back. On the way out it suffered axle-box problems; on the way back, a turbine failed. It never ran again.

  The idea of the steam-turbine electric locomotive continued to inspire railway engineers. The most ambitious experiments, all of them failures, were made in the United States between 1938 and 1954, by the Union Pacific Railroad, the Chesapeake and Ohio Railway, and the Norfolk and Western Railway. The feeling was that the new diesel-electrics might just be rivalled by a new generation of steam-turbine electrics which not only looked like diesels but had similar speed and tractive effort characteristics. In April 1939, at the instigation of Otto Jabelmann, General Electric delivered a pair of striking 277 ton, vacuum-condensing, oil-fired, steam-turbine electric 4-6-6-4s (or 2-C + C-2s) to the Union Pacific; the railroad returned this twin 5,000 hp machine to the manufacturers for modifications just two months later. Encased in a long-nosed, diesel-style casing, each oil-fired 2,500 hp unit featured a semi-flash 1,500 psi water-tube boiler for rapid steam raising, high- and low-pressure turbines, two electric generators, six air-cooled traction motors, and dynamic electric braking. After further trials on the New York Central Railroad and the Great Northern Railway, the twins were retired in late 1943; wartime conditions were not conducive to prototype locomotive development.

  In 1946, the Chesapeake and Ohio Railway proposed to introduce a high-speed ‘Chessie’ passenger service between Washington, DC, and Cincinatti. High power would be required for climbs through and over the Alleghenies. Mainly for this service, the railway ordered three 6,000 hp, non-condensing, steam turbo-electrics. Built by Baldwin with Westinghouse Electric in 1947–8, these 154 ft, multi-axle machines, weighing 367 tons, promised 6,000 hp and 100 mph, and a glimpse into the future. Their large 310 psi fire-tube boilers, similar to those of the Pennsy’s S2 class 6-8-6, supplied steam to a Westinghouse impulse turbine working in its economic range between 3,600 and 6,000 rpm and driving a pair of twin-armature generators. While all this sounded impressive, there was a fundamental flaw with the design. The turbine designers had not been told by Baldwin that the turbine would be exhausting against a 25 psi back-pressure to draught the boiler. They had assumed that the exhaust would be at atmospheric pressure. The result of this misunderstanding was a reduction in maximum turbine power output of 20 per cent. Dr Adolph Giesl-Gieslingen, the Austrian engineer and inventor of the Giesl exhaust ejector, who had already fitted one of these devices to a Chesapeake and Ohio 0-8-0 switcher (shunter), designed a special twin ejector for the turbo-electrics to reduce back-pressure enough to raise maximum power to 5,600hp. It was, however, too late. The new Chesapeake and Ohio president, Walter J. Touhy, cancelled the ‘Chessie’ high-speed train project and the Giesl ejectors were never fitted to the other turbo-electrics. The locomotives were put to work on heavy freight trains instead, and withdrawn in 1951.

  One final attempt was made by the Norfolk and Western Railway in 1954 with 2300 Jawn Henry. Named after the legendary construction worker, John Henry, who took up a challenge to outperform a steam-hammer and won, only to collapse and die, this formidable-looking, glossy black, 4,500 hp, non-condensing machine, with its Babcock & Wilcox 600 psi water-tube boiler, twelve electric traction motors, and 6-6-6-6 (or C-C-C-C) wheel arrangement, was itself the stuff of legend. Jawn Henry could move a heavy coal train away from rest with greater aplomb than one of the railway’s magnificent Y6b Mallets. ‘Jawn rides smoothly and makes the dynamometer car needles move with gusto,’ reported David Morgan, editor of Trains magazine, in May 1953. ‘Once while starting, the tractive effort measurement read 224,000 pounds. Impressive!’ Impressive indeed: a Y6b had a maximum tractive effort of 152,000 lb. In tests over mountain sections in 1954, 2300 hauled loads on average 18.27 per cent heavier than the Y6bs, and although at speeds 4 to 13 per cent lower, these runs were achieved with fuel savings over the Mallets of between 20 and 30 per cent.

  Following these tests, the Norfolk and Western decided to purchase up to twenty further turbo-electrics, with power raised to 5,250 hp. Unfortunately, this decision coincided with the crisis leading to the withdrawal of Baldwin from locomotive building and the arrival of the new Norfolk and Western president, Stuart Saunders, who was determined to abandon steam in any form in favour of diesel-electrics. This was especially sad as the Norfolk and Western test report on Jawn Henry, published on 26 April 1955, had stated: ‘There is a definite place and need for a coal-burning locomotive on some American railroads. The advantages of the turbine-driven over the reciprocating engine are being recognized. Further development of the oil-burning
turbine-electric locomotive is assured. Other railroads are interested. The 2300 points the way.’

  Jawn Henry was a descendant of the intriguing Heilmann steam-electric locomotives invented by Alsace-born Jean-Jacques Heilmann, proprietor of the Le Havre-based Société Industrielle de Moteurs Électriques et à Vapeur, in 1890. These were the first self-propelled locomotives in which the prime mover (steam, diesel, or gas turbine) operated in its efficient speed range with electrical transmission providing a variable speed drive to the road wheels. Heilmann’s first steam-electric locomotive was La Fusée Électrique (The Electric Rocket), built in Paris in 1892–3 in collaboration with the Winterthur locomotive works and Bown, Boveri & Cie. A cab-in-front design, the 118 ton machine had a special Lentz boiler with a stay-less corrugated circular fire-box. Steam passed to a twin opposed-cylinder, 790 hp, compound engine, driving a 400 kW DC generator which, in turn, powered eight 80 hp traction motors, one mounted on each of the locomotive’s eight axles, carried in two bogies.

  On a trial run with the Ouest railway on the Paris to Nantes main line, La Fusée Électrique covered 58.33 kilometres (36.25 miles) of undulating line with a load of 183 tons at an average speed of 70.4 kph (43.75 mph), topping 100 kph (62 mph). Two more powerful, 124 ton, 1,350 hp locomotives, with conventional fire-tube boilers and Willans six-cylinder triple-expansion engines, were built in 1897–8. On test, the first of these, No. 8001, ran happily at 120 kph (74.5 mph) with loads of 250 tons. With 100 per cent adhesion – the engine’s entire weight rested on its driving wheels – good acceleration, and quiet running, the Heilmann locomotives were a real success. They accelerated strongly and worked on the Paris–Versailles suburban service until the line was electrified. The Ouest railway, however, required that they carry a crew of three – driver, fireman, and electrician – a fact that did nothing to help their cause, while their power-to-weight ratio was low compared with that of the best new French compound Atlantics.