Giants of Steam

by Jonathan Glancey

  Intriguingly, the idea of a heavy-duty American reciprocating steam-electric locomotive was revived in the 1990s in a project by two British engineers, Robin Comyns-Carr and George Carpenter. Designed for heavy freight work, this 4,500 hp, vacuum-condensing Co-Co locomotive was to have used a coil-tube, semi-flash boiler working at 1,700 psi, feeding a twelve-cylinder triple-expansion engine, driving electrical transmission.

  Attempts to combine the reciprocating steam engine with electric power were also tried at various times, with varying degrees of failure and success. One unusual development was the conversion during the Second World War of several veteran Swiss federal railways 0-6-0 tank engines into steam-electric locomotives. Fitted with pantographs, the E3/3 class locomotives drew power from 15 kV overhead lines. This fed electric heating elements in the boiler through transformers rated at 480 kW. At a time when Switzerland was desperately short of coal – it has no indigenous supply – yet rich in hydroelectricity, it seemed only sensible to power steam locomotives in this way. The E3/3s could be coal-powered when necessary. The idea was a wartime expedient, though; in Swiss conditions, ultimately it made more sense simply to build electric locomotives.

  *

  Steam-diesel hybrids, meanwhile, were investigated by railways in Britain, Russia, and Italy. As both forms of traction were seen to have their advantages, perhaps a combination of the two might be a winner, with steam providing power at starting and thus making it unnecessary to equip diesel engines with variable-speed transmission. The idea was patented in 1917 by William Joseph Still, a London engineer, who presented its virtues at a meeting of the Royal Society of Arts chaired by Charles Algernon Parsons. Still’s invention was a reciprocating engine with double-acting cylinders. The top half of the cylinder was diesel, the bottom half steam. At starting, steam was supplied from an oil-fired boiler. Once running, it was generated by hot diesel exhaust gas and steam propulsion could be used as a booster to the diesel cylinders when necessary.

  The first Still engines were used to power ships. In 1924, Dolius, an 11,533 ton Blue Funnel Line freighter, built by Scott’s of Greenock, set off on her maiden voyage from Cardiff to Algiers. She averaged 11.45 knots and burned a total of 8.4 tons of fuel per day. This was widely noted in the shipping industry: it was a fine and economical performance. But the ship required engineers trained for both steam and internal-combustion working.

  The first such engine on rails was the Kitson-Still locomotive, a 2-6-2T built by Kitson & Company of Leeds, for trials on the LNER in 1926. The cylinders sat below the small, high-pitched, oil-exhaust-gas-fired boiler. The engine unit had two horizontally opposed banks of four cylinders on either side of the crankshaft, in parallel with the driving axles. The 85 ton oil-fired engine had 5 ft 0 in driving wheels, a 200 psi boiler pressure, and a tractive effort of 28,000 lb. Starting under steam, diesel power cut in at 5 mph and the locomotive could run up to 43 mph at a maximum engine speed of 450 rpm.

  The Kitson-Still machine proved to be a strong starter and could manage 400 ton goods trains, restarting these on gradients as steep as 1-in-50. Maximum diesel power was 800 hp, with up to about 200 hp more when boosted by steam, although 80 to 90 per cent of running was diesel only. Fuel consumption compared well with much later British Railways main-line diesel locomotives. But it was no match for Gresley’s conventional steam locomotives, and because great strides had been made with electric transmission for diesel engines in the course of the 1930s, the justification for the Kitson-Still engine was diminished. A fascinating and not unsuccessful experiment, the locomotive was broken up by Kitson and Company in late 1935.

  Russia was one of the very few countries (another was Italy) to pursue the idea of a steam-diesel locomotive. The Teplopaorvozi (diesel-steam) locomotive was built at Voroshilovgrad in 1939. A hunched and massive-looking 2-8-2, No. 8000 was equipped with four opposed pistons mounted centrally outside its frames, each pair driving the coupled wheels through connecting rods and jackshafts. Up to 20 kph (12.5 mph), steam acted on the outside of the pistons and internal combustion on the inner. Above this speed, internal combustion took the dominant role, although steam continued to drive the outer piston faces on their inward movements. Its smoke-box door emblazoned with the legend ‘Stalinets’ (‘Follower of Stalin’), 8000 was meant to have produced 3,000 ihp and run at up to 130 kph (81 mph). Trials began in October 1939, when 2,950 ihp was developed at 78 kph (48.5 mph), and continued until 1943. Although testing resumed in 1946 after the Great Patriotic War of 1941–5, the locomotive was stored two years later and preserved. Apparently it suffered from cracked cylinders and a poor ride, and with its 25 ton axle loading it was in any case too heavy for almost all Russian track.

  A second, quite remarkable, Soviet steam/internal-combustion locomotive made its unlikely debut from Kolomna works on 26 December 1939. This was the TP-1 2-10-2, a cab-in-front design with opposed pistons, powered by pulverized and gasified anthracite for the internal-combustion cycle, and by steam raised from the pulverized coal left over from the gas-producing cycle. This was a complex process, made more intricate and demanding still by the fact that this was a condensing locomotive complete with a turbine exhaust fan. The driver sat in the diesel-style enclosed cab at the front of the dark-blue and red engine, while the fireman regulated steam and gas production from a conventional steam locomotive footplate – although, this being Russia, one fully protected from the elements.

  The valve gear was immensely complex and, in action, must have looked rather like a spider cycling. Mounted more or less in the centre of the frames, TP-1’s eight pistons drove the engine’s wheels through four crossheads on each side, via jackshafts and rocking levers. This bizarre machine was intended to match the performance of a highly competent FD class 2-10-2. The gas cycle was to have produced 2,000 ihp and the steam cycle a further 1,000 to 1,500 ihp. In the event, TP-1 made seventy-six runs around the Shcherbinka test track, covering a total of 1,790 kilometres (1,112.25 miles). It seems that it did work as planned, but only up to about 30 kph (18.5 mph). Above this speed, the gas cycle gave up the ghost. Kolomna engineers worked hard to improve TP-1’s performance, but again the war intervened and there was no time, let alone will, to continue such experiments. Later, in 1948, a third opposed-piston steam/ internal-combustion locomotive was built, to a pre-war design, at Voroshilovgrad, but although this eight-cylinder 2-10-2, numbered 8001 and weighing 153 tons, was less complex than TP-1, it was considered to be a complete failure.

  The idea of using pulverized fuel, however, had also interested German engineers in the lead-up to the Second World War, especially as the country had large deposits of lignite. The Deutsche Reichsbahn’s 05 003 4-6-4 was the third of this small class of high-speed streamliners. Here, the conventional layout of boiler barrel and fire-box was reversed, with the fire-box at the front immediately behind the driver’s cab. The entire machine was encased so that when steam was shut off it could easily have been mistaken for a new diesel or electric locomotive. With the fire-box remote from the tender, the grate was fed by pulverized coal blown through delivery pipes by a centrifugal compressor. Unlike its super-fast siblings, though, and despite its promising looks, 05 003 was a relative failure; it was converted into a conventionally arranged and fired 4-6-4 by Adolf Wolff, Krauss Maffei’s post-war technical director, in 1950.

  Attempts to improve the way in which power was delivered – ways, that is, of reducing torque fluctuation when transmitting cylinder thrust to driving wheels – led to a number of fascinating experiments in the 1930s with multi-cylinder engine units designed to power two or more axles and thus adopt one of the great advantages of diesel engines: smoothly transmitted power, especially from starting. In France in 1939, the SNCF’s Vitry test plant produced the solitary 221TQ, a 1,200 hp 4-4-2 tank engine. This boasted a high-speed steam V12 engine, mounted longitudinally, driving two axles through a transmission shaft and helical gears. The prototype had been ordered by Dabeg, manufacturers of locomotive poppet valve
s and feedwater heaters, and was intended for secondary-line service. The war put a stop to its development and, when in 1946, the SNCF announced an end to all unconventional types, 221TQ was scrapped.

  In July 1941, however, V19 001, the eight-cylinder 2-8-2 masterminded by Friedrich Witte, head of testing for Deutsche Reichsbahn, with Richard Roosen of Henschel, began trials. Streamlined in the same style as the record-breaking 05 Baltics, V19 001 was intended for speeds of up to 175 kph (109 mph). Each axle was powered by a V2 engine. The war put an abrupt end to this promising experiment. The Witte-Roosen ‘V8’ certainly looked the part but, disappointingly, steam consumption rates were higher than those for the 05s. The locomotive, however, drew considerable attention from US Army Transportation Corps engineers – perhaps the idea of a ‘V8’ engine appealed – and V19 001 was shipped to the United States. After enduring years of neglect, she was scrapped in 1957.

  *

  Certain features developed through some of these experimental designs did meet with wider acceptance, although these were often the simplest elements. From 1910, cab-in-front design, for example, revolutionized steam operation on the Southern Pacific Railroad’s demanding route through the Sierra Nevada. Here, trains ran through miles of tunnels and snow sheds on a steeply inclined line. Working through these with a conventionally arranged steam locomotive, with fumes and excessively high cab temperatures (which could rise to 60–70 °C), was not just unpleasant but downright dangerous. The Southern Pacific’s two hundred or more oil-fired, cab-in-front 4-8-8-2s, built up until 1944, solved the problem. The crew worked from a cab at the front of these powerful and greatly respected locomotives, with the fire-box behind them; this gave them a clear view of the twisting, climbing, snow-covered track ahead, and ensured they were not troubled by fumes because the chimney, a long way behind, faced the tender and exhaust was carried back over the train.

  Equally successful were the many Shay-geared locomotives built by Lima over the decades from 1880. These delightful and highly effective engines were designed for use on lightly laid, often narrow-gauge, railways, where very steep and slippery gradients and tight curves were the norm. Employed widely in the logging industry, some even ran on wooden rails. To ensure optimum adhesion and the smooth delivery of power, albeit at low speeds, the Shay locomotives were fitted with three vertical cylinders mounted on one side, driving a longitudinal shaft (as in a lorry or car), geared to the road axles of the bogies they rode on.

  What became, however, one of the holy grails of some steam locomotive designers in the mid-twentieth century was the idea of a modern main-line steam locomotive with mechanical drive and the total adhesion enjoyed by double-bogie diesels and electrics. Such a machine would have a cab at both ends. It would be fast, powerful, and economical to run, with excellent acceleration. Sheathed in sheet-steel casing, it would be easy to clean. It would make turntables redundant. It would not just prolong, but might even revolutionize, the story of the steam railway locomotive.

  As it was, the first locomotives of this type were the metre-gauge Co-Cos built by Sentinel in 1933 for Colombia’s railways, with Woolnough semi-flash boilers working at 550 psi and two-cylinder compound engine units driving each axle. One engineer who was attracted to the concept, and who clung to it with some tenacity, was Oliver Bulleid, when he was chief mechanical engineer of the Southern Railway and, later, of the CIE in Ireland. The locomotive he developed, in the face of mounting opposition within his own mechanical engineering department on the Southern Railway and, from 1948, the management of the new British Railways, was the legendary Leader 0-6-6-0T of 1949. Here was a steam locomotive that really could be mistaken for a diesel.

  Curiously, the Leader project began, in 1944, with a request by the Southern Railway’s general manager for proposals for twenty-five general-purpose tank engines. This gradually developed into an all-purpose locomotive that would modernize locomotive operation on the non-electrified sections of the Southern Railway. Indeed, in a paper published in Modern Transport on 26 October 1947, entitled ‘New Conception of Steam Locomotive Design – Lessons for the Immediate Future’, Bulleid asked: ‘What sort of locomotive may we expect to see if it is to meet the majority of our future requirements?’ Answering his own question, he wrote: ‘The locomotive should be built (1) to run over the majority of company lines; (2) to be capable of working all classes of train up to 90 mph; (3) to have its whole weight available for braking and the highest percentage thereof for adhesion; (4) to be equally suited to running in both directions without turning with unobstructed look-out; (5) to be ready for service at short notice; (6) to be almost continually available; (7) to be suited for complete “common use”; (8) to run not less than 10,000 miles between general overhauls with little or no attention at the running sheds; (9) to cause minimum wear and tear to the track; and (10) to use substantially less fuel and water per drawbar horsepower developed [than conventional steam locomotives].’ (As it turned out, of these desiderata only numbers 3 and 4 were fully achieved.)

  With these points in mind, Bulleid declared that ‘a new type of Southern engine has been designed, the construction of five has been authorized. The engine will incorporate the following features . . . The locomotive is carried on two six-wheeled bogies, the general design of which follows that of the bogies I designed for use under the company’s electric locomotives [they were, in fact, designed by the Southern Railway’s Percy Bolland] . . . the engine develops a torque, the uniformity of which is comparable with that of a nose-suspended electric traction motor but has a higher speed range and the unsprung weight is less. The capacity of the boiler has been made greater, relative to the cylinder horsepower, than in the case of any previous Southern locomotive. The cabs at both ends will give an improved lookout . . . The engines are intended for working fast passenger trains of 480 tons weight over the difficult Southern main line, and goods and mineral trains of up to 1,200 tons; that is to say, something above the heaviest trains that would be required on the system.’

  Material for the first five Leaders had been ordered in December 1946. The locomotive, 36001, that emerged from the Brighton works, painted a uniform grey, was certainly like no steam locomotive that had ever been seen before. It promised a great deal, but in the event proved to be a troublesome prototype which was attempting to apply several undeveloped design features. The British Railways locomotive design team, led by E. S. Cox under ‘Robin’ Riddles, steeped in the ideas of Stanier, standardization, austerity, and simplicity, were never going to be sympathetic to such a radical and complex machine unless its new features were quickly proven to be reliable – particularly since the Stanier class 5 4-6-0, and later the British Railways Standard class 5 4-6-0, could do pretty much anything the Leader could do, with exemplary reliability and economy.

  ‘As a man, I liked Bulleid and we got on well,’ said Riddles after the latter’s death. ‘But he was an individualist who wanted to do it his own way and some of what he did was questionable. I couldn’t fathom some of his thinking . . . Was there a job for Bullied after Nationalization? No; how could there be? Having been king of his own dung heap, how could he suddenly come and serve in mine? The surprise to me was that having failed on the Southern, he went to Ireland and failed there too.’

  This uncharacteristically tart comment by the generally broad-minded Riddles shows just how strongly more conventionally inclined engineers than Bulleid felt about the Leader and its successor, the Irish Railways’ turf-burning locomotive, CC1, built at Inchicore in 1957. And sadly, the age of austerity, when the British public still shopped with wartime ration books and national finances were very tight indeed, meant that there was little room for prolonged experiments like the Leader, which had no guarantee of success and was paid for by the public purse.

  Although 36001 had been retired in March 1951 and construction of the remaining four Leaders halted, the Sunday Dispatch nonetheless led its 18 January 1953 edition with the headline ‘Railways’ Biggest Fiasco’. The pop
ular Sunday newspaper had sent its reporters to investigate the scandal of ‘£500,000 being wasted on three useless engines’. ‘The engines, 67-foot-long monsters of the Leader class with driving cabins at each end, were built in 1948 and 1949 to “revolutionize” train travel. They were part of a £750,000 experiment undertaken by Mr O. V. Bulleid, Chief Mechanical Engineer of British Railways, Southern Region. I can reveal,’ thundered the anonymous Dispatch reporter, ‘that the region’s officials regard the experiment as one of their biggest failures.’

  Other stories on the front page of that edition of the Dispatch told of a twenty-year-old from Glasgow winning the first heat of the National Bathing Beauty Championship and of Manny Shinwell, ‘a former Socialist Minister for Defence and Secretary of State for War, attacking plans for Queen Elizabeth’s Coronation because standard bearers at this great state occasion were to be drawn from the aristocracy and particularly from the military’. And, at the bottom of the page, a small headline reading ‘Fares Rise Now May Be Queried at Westminster’ gave the game away. Fares rises were unpopular then, as now, and so British Railways could be taken to task for demanding higher revenues while throwing money away on ‘white elephants’.

  The problem with the Leader was that too many experiments were carried out on one locomotive at one time and with inadequately proven components. The three-cylinder bogies had chain drives coupling the centre driving axle with the outer axles. The valve gear was housed in an oil bath which was liable to leak, and sleeve valves, more often associated with car engines than steam railway locomotives, were subject to steam leakage and seating-ring breakage. The stubby 280 psi welded boiler was offset to one side of the locomotive’s frames to allow for a passageway giving access to the fireman’s compartment. Furthermore, since the chosen fuel for the Leader was coal rather than oil, the fireman was tucked into a tiny and blisteringly hot cubbyhole, nicknamed ‘The Chinese Laundry’, inside the engine’s casing. And because of the nature of the fire-box, in which the sides were provided by firebrick walls which needed thickening in service, the grate area was just 26 sq ft, rather than the 43 sq ft originally planned. Meanwhile, just 2,500 gallons of water were carried, which meant that the locomotive needed to make frequent stops to replenish her tanks. The weight grew during the design and construction stage from 110 to 130.5 tons – much of this due to the ballast needed to balance the offset boiler – and, given its axle loading of 21.75 tons, the chief civil engineer would only allow 36001 to run on lines cleared for the Merchant Navy Pacifics and, initially, at a maximum of 50 mph.