Giants of Steam

by Jonathan Glancey

  He had. Chapelon had made a great leap forward in the design of the steam locomotive. On a power-to-weight basis, 3566 was the world’s most powerful steam locomotive. Its performance was no flash in the pan: the engine could be worked hard, when necessary, for mile after mile, day after day. And this was just the beginning. The other French railways began following Chapelon’s principles and rebuilding their Pacifics and other passenger locomotives, especially after trials held on the Nord railway in the winter of 1932–3 which set Chapelon’s 3566 against a Nord railway Super Pacific of 1931, an Est railway 4-8-2 of 1925, and the PLM’s latest 4-8-2, also of 1931. Regular trains worked included the heavy Pullman Flèche d’Or on its Calais Maritime to Paris Gare du Nord run. The Nord railway inspectors were suitably impressed when 3566 hauled the 650 ton Pullman up the 24.9 miles of the Gannes bank, a gradient of between 1-in-333 and 1-in-250, at an unflinching 120 kph (74.5 mph).

  The Nord railway ordered twenty Paris–Orléans rebuilds, as well as twenty-eight brand-new Chapelon Pacifics built by private industry. These became the 3,400 ihp 231E class and could still be seen at the head of the Flèche d’Or in the mid 1960s, still turning in the same precise performances as had been expected of them three decades earlier. The last was withdrawn in 1967. Weighing just 91 tons, or 160 tons with tender, the new Chapelon Pacifics had a power-to-weight ratio not just higher than that of any other steam locomotive of the time, but higher than that of some much-vaunted diesels, like the English Electric Deltics, the class of diesel-electrics designed to take over from Gresley Pacifics on the east coast main line in the early 1960s. The Deltics had a maximum power output of 3,300 ihp – and while this could be produced at the turn of a handle, much the same was true for the French Pacifics.

  The new Nord Pacifics could have been more powerful still. In 1932, Chapelon converted one of the smaller-wheeled saturated Paris–Orléans 4500 class Pacifics of 1907 into a four-cylinder compound 4-8-0. This was done to improve performance on the southern stretch of the 443 mile main line from Paris to Toulouse. The last 250 miles were characterized by steep gradients and continuous reverse curves. Many trains had to be double-headed, a policy that was largely anathema in cost-conscious France. What is more, the Paris–Orléans railway wanted to accelerate passenger trains on this route, especially during the popular skiing season in the Pyrenees, and to increase their maximum weight from 500 tons, to which the Pacifics were limited by considerations of adhesion, to eighteen-coach loads of 700 tons. The result was 4521. Again, this was a rebuild of an earlier Pacific, one of the 4500 class of 1907, the very first European Pacifics.

  With extended mainframes, a new long, narrow fire-box with a 40.5 sq ft grate, a boiler pressure of 290 psi, and Chapelon’s thermodynamic elixir, the reborn 4-8-0 was the prototype for one of the greatest of all classes of locomotives at work anywhere in the world, the 240P of 1940. The eight-coupled, four-cylinder compound 4521 emerged in steam from the Tours works on 16 August 1932. The 104.5 ton locomotive’s performance was truly sensational. She soon proved capable of maintaining 3,450 ihp continuously, with a maximum of 4,000 ihp, and a maximum recorded speed of 94 mph, considered very fast indeed at the time for a locomotive with 1.85 m (6 ft 0¾ in) driving wheels. On one test, 3,030 dbhp was sustained for one hour at 62 mph, during which time the fireman, chauffeur Marty, fired four tons of coal. He had previously been a torpedo-boat stoker, but said that he preferred working in the open air.

  Equally at home at the head of heavy freight and express passenger trains, 4521 created the greatest interest among train operators throughout France. Twelve of these Chapelon 4-8-0s were built. Two were loaned to different French railways, and their performance on test never failed to impress. On 18 February 1935, 4707, now renumbered 240.707, ran a 650 ton test train over the 183 miles from Calais to Paris, with a water stop at Amiens, at an average running speed of 71.2 mph, and averaging 87 mph over 20 miles of level track. On 21 March, the Deutsche Reichsbahn engineers Drs Fuchs and Nordmann rode a fourteen-coach, 607 ton test train on the État railway between Paris and Cherbourg – 230 miles over heavy inclines – behind the same engine. Both men were deeply impressed as they saw it achieve 2,800 dbhp on the banks, equivalent to 3,200 dbhp on level track, something they had thought impossible from such a compact and – to German eyes – complex-looking machine. Back in Berlin, a proposal for a Deutsche Reichsbahn 4-8-0 derived from the Paris–Orléans machine was prepared at the Borsig works, although the war put a stop to this Franco-German design.

  On loan to the PLM – by now the SNCF’s south-east region – in 1938, 240.705 worked the fastest expresses over the 317.5 miles from Paris to Lyons, gaining up to half an hour on schedules booked at more than a mile-a-minute. Marcel Japiot, the PLM’s new chief mechanical engineer, was suitably impressed – and even more so when, on 18 July, 240.705 was put at the head of a fifteen-coach express weighing 672 tons, more than double the normal 320 tons of the Côte d’Azur Pullman. Ten minutes late away from Laroche, through no fault of its own, the Chapelon 4-8-0 was on time at Ancy le Franc forty miles on, and five minutes ahead of time at Blaisy summit, having gained 15 minutes in 83 miles.

  This was in 1938, the year the French railways were nationalized. Japiot, who was now chief mechanical engineer of the south-east region, had little hesitation in ordering twenty-five 4-8-0 rebuilds of the Paris–Orléans 4500 class Pacifics. Significantly, Japiot had succeeded René Vallentin, who resented Chapelon’s achievements, especially when, in 1936, 240.705 outperformed 241.C1, the PLM’s prototype large-wheeled 4-8-2, in terms of both greater power output and lower fuel and water consumption.

  Chapelon took the opportunity further to strengthen the frames at the rear of the locomotives, slightly to enlarge the inside low-pressure cylinders, and to install mechanical stokers. Justly proud of the new engines – the 240P class – Chapelon wanted to improve their appearance over the dozen Paris–Orléans 4700 class 4-8-0s. He asked his friend, the artist Émile André Schefer, to help with their styling. Schefer’s inspired paintings of French railway locomotives and posters for the railway companies, capturing and framing images of speed, grace, and power, were well known and very popular. The result was the raising of the running board clear of the driving wheels, for greater accessibility, a larger boiler casing, under which most of the external pipework of the earlier 4-8-0s was concealed, a PLM-style ‘windcutter’ cab, and a conical smoke-box door, complete with Schefer’s version of the SNCF’s famous circular badge. The result was one of the best-looking of all French locomotives, as well as the steam locomotive with the highest power-to-weight ratio anywhere in the world. The fact that no 240P has survived into preservation is little short of a tragedy.

  The 240Ps were rebuilt from Paris–Orléans un-superheated Pacifics at Tours and Périgueux between May 1940 and October 1941. In June 1940, the French had surrendered to the Germans. The armistice was signed in the Compiègne forest inside carriage 2419 of the Orient Express, a handsome restaurant car brought from a museum in Paris for the occasion – the Germans had been forced to sign the armistice in the very same coach in the exact same location on 11 November 1918. The Germans imposed a national railway speed limit of 100 kph (62 mph), although speeds had already been cut by the SNCF after the declaration of war in order to minimize wear on track and rolling stock, so the new 240Ps were not required to run fast. Their power, however, proved to be prodigious.

  Testing began in spring 1941. On 31 May 1941, 240P.5, with driver Chartier and fireman Jarry of Laroche on the footplate, set a new world record for steam power. From Laroche, and with seventeen cars weighing 800 tons behind the tender, the compound climbed the 19.4 mile gradient, averaging 1-in-187 and peaking at 1-in-125, between Les Laumes and Blaisy without the speed falling below 60 mph. Power increased steadily as the gradient steepened, with 240P.5 maintaining full boiler pressure, reaching 4,400 ihp, equivalent to 3,600 dbhp – very nearly 40 ihp per ton of locomotive weight. This was greater than the power-to-weight ratio n
ot only of any other steam locomotive but also of any diesel locomotive of the next decade. Given that 240P.5 was mechanically fired, there had been no need for the heroics of chauffeur Marty. If Chapelon had ‘done something’ with his first Pacific, he had done something else again with the 240Ps.

  The 240Ps were mainly shedded at Laroche, working to Dijon, Lyon, and through to Paris. At the end of the Second World War, they headed the sole main-line day passenger train from Paris to the south of France, loaded with up to twenty-eight coaches, or 1,400 tons. And yet, these herculean 4-8-0s were broken up by the end of 1953. This seems extraordinary, as surely these young, immensely powerful, and hugely competent locomotives – Chapelon’s favourites among his own engines – should have been transferred elsewhere on the SNCF. But unfortunately, Chapelon – a kind and profoundly religious man with no ego and no side to him – was resented by some in management for continuing to argue that there was still a place for steam in the SNCF’s post-war motive power policy. He was increasingly regarded by these people as a passéiste, a man of the past. His steam locomotives were resented too. Even so, there were plans to move the 240Ps to the south of France, where they would have radically improved services over the difficult, if often beautiful, coastal route to Nice. However, this would have required conversion to oil-firing because of the threat of line-side fires posed by coal-fired locomotives in this hot, dry, inflammable region, and the 240Ps were rejected because the SNCF had ruled that no further development could be made to steam locomotives after 1951.

  Chapelon believed that there was always considerable resistance to his proposals for new and rebuilt locomotives from forces within the motive power running departments on the Paris–Orléans railway. (In France such departments came under the aegis of the chief mechanical engineer until 1970.) This was overcome by the determination of the Paris–Orléans’s chief mechanical engineer Édouard Épinay and his chief engineer for rolling stock and works, Marcel Bloch. On the Est and within the SNCF, there was similar resistance.

  Chapelon was appointed a chevalier of the Légion d’honneur in 1934 for his work in locomotive development and won increasing fame outside of France, especially after the publication of La Locomotive à Vapeur in 1938. But he rose only slowly up the railway hierarchy. A dedicated design engineer and not a management man, he was eventually appointed the SNCF’s chief steam locomotive design engineer, although he was never, like so many of his disciples around the world, a chief mechanical engineer. He did, though, have a remit to produce new designs both during and after the Second World War, including designs for an altogether new generation of high-power steam locomotives which would make even the 240Ps seem underpowered.

  Chapelon’s most innovative design was a prototype six-cylinder compound 2-12-0. Built at Tours in 1940, 160.A1 was kept hidden from the Germans and was only fully tested from 1948. The idea was to shape a locomotive capable of hauling heavy freight trains at slow to moderate speeds over extremely hilly routes with the thermal efficiency of an express passenger engine working at speed. The idea had originated as early as 1933, when the operating department of the Paris–Orléans had been looking for a locomotive to haul 1,200 ton freight trains over the long 1-in-100 gradients of the steam-operated section of the Toulouse line. Chapelon believed he could produce a 2-12-0 with half as much power again as the existing four-cylinder compound 2-10-0s working these trains. Although commissioned in 1935, it was five years before 160.A1 was first steamed and another year before she was able to prove her remarkable power and efficiency. At the time of the German invasion, the brand new locomotive was returned to her home depot, Brive, at the head of a 1,200 ton freight train, without having been run in; she climbed the heavy gradients from Limoges at a steady 40 kph (25 mph), producing a sustained 2,700 dhp in the process.

  Steam locomotives are generally more efficient the higher their rotational speed, measured in rpm. At lower rpm and piston speeds, more heat is lost through the cylinder walls, condensation, and leakages. To minimize these, Chapelon used steam-jackets surrounding the cylinders between the high-pressure and low-pressure stages. Four low-pressure cylinders gave the increased cylinder volume required to enable steam to be used expansively at low speeds, employing relatively short low-pressure cut-offs.

  Tests conducted after the war, in May and June 1948, vindicated Chapelon’s ideas. The engine was easy on coal and water and was powerful throughout its speed range, peaking at 2,750 dbhp – sustained on test over three hours without a break – at 60 kph (37 mph) with a train of 1,650 tons. She used 40 per cent less coal than a three-cylinder 150X class 2-10-0 (German class 44s of the SNCF) when both locomotives were worked at 1,600 dbhp.

  Further tests demonstrated that steam-jackets eliminated the need for superheating the high-pressure cylinder and for very high initial steam temperatures, superheating being provided only between the high-pressure and low-pressure stages. Sadly, the 2-12-0 was taken over by events as the Brive to Montauban section of the Toulouse line for which it had first been planned, fifteen years earlier, had been electrified in 1943. It was also too late to affect the design of the last compounds built for the SNCF, although its principles were to have been applied to Chapelon’s proposed, if ill-fated, high-power designs. Although it should have been preserved, 160.A1 was scrapped in 1955.

  During the war, Chapelon not only pushed on with wholly new designs but was also partly responsible for 318 mixed-traffic 2-8-2s. These were the 141P class four-cylinder compounds of 1941, designed in collaboration with Georges Chan, an outstandingly powerful and reliable type, with 1.7 m (5 ft 5 in) driving wheels, which could run up to 110 kph (68 mph) on test and could sustain 3,330 dbhp at 80 kph (50 mph). The maximum power of the 141Ps, at 4,000 ihp, was almost equal to that of the Chapelon 4-8-0s at speeds up to 100 kph, and double that of the PLM Mikados they replaced.

  At the end of the war, SNCF ordered 1,340 simple-expansion two-cylinder 2-8-2s from the United States as a matter of urgency to help replace the thousands of locomotives destroyed between 1939 and 1945. When the SNCF was formed in 1938, it had a fleet of approximately 17,800 steam locomotives; by May 1945, this had been cut by German appropriation to 12,000, of which only about 3,000 were in running order. The new 141R class, of which 1,323 went into service, the other seventeen being lost off the coast of Newfoundland on the way to France in April 1947, were very different from the 141Ps.

  Featuring robust frames, roller bearings, Boxpok disc wheels, and much larger cabs than French crews were used to, these rugged Jeeps of the French railways were based on a 1937 Alco design for the Green Bay and Western Railroad and were put into common-user service, meaning that, unlike their more evolved French cousins, they were driven and fired by any number of crews. Extremely reliable – they ran up to 125,000 miles between general overhauls – they were popular and remained at work up until the end of steam on the SNCF in 1974. They were neither as fast – limited to 100 kph – nor as powerful – 2,633 dbhp at 50 mph – as the 141Ps, but they did show the value of rugged American engineering, and the strides that had been made across the Atlantic in terms of new manufacturing techniques which made engines easier and quicker to build than in much of Europe, as well as inherently more robust.

  The second batch of 141Rs consisted of 640 locomotives fitted with single Kylchap exhausts, boosting their maximum sustained power output to 2,928 dbhp at 50 mph. The first batch used 40 per cent more coal and water than the ex-PLM Mikados but were cheaper to maintain. Before the end of SNCF steam development in 1951, Chapelon was able to modify 100 141Rs with improved valves and steam passages and higher superheating, improving fuel economy by 15 per cent and raising their maximum power to 3,300 dbhp. For many holidaymakers to the south of France into the early 1970s, the 141Rs were their first and abiding impression of French steam, as they worked smartly timed passenger trains along the Côte d’Azur from Marseilles to the Italian border.

  Another of Chapelon’s wartime designs had been for modular utility locomotive
s. These were based on a two-cylinder simple-expansion engine which could be stretched from a 2-6-0 to a 2-8-0 or even a 2-10-0. With 1.7 m (5 ft 5 in) driving wheels, all three classes would have been able to run at up to 110 kph (68 mph). These would have been successful machines, but they proved unnecessary since by the time they would have been ready for production, North American locomotive works were already making 141Rs at a rate quite beyond the capacity of the domestic French industry.

  Like so many French people at the time, Chapelon coped with the Nazi occupation as best he could. Soon after the war’s end, however, he was one of those disgracefully arraigned by an SNCF ‘purification’ committee, which accused him of demonstrating a ‘passive attitude towards the enemy’. He was to be questioned immediately after Marc de Caso, who, like Chapelon, was a decorated First World War artillery officer and former student of the École Centrale, and had gone on to join the Nord railway and to design a number of impressive, efficient, and beautiful 4-6-4s at the beginning of the war.

  Born in Montepellier to a family with Italian and French ancestry, de Caso was an ebulliently outspoken and passionate man, and also a great admirer of Chapelon. He liked to say: ‘I don’t leave my tongue in my pocket.’ As he lambasted the committee, one of whom was a railway engineer who resented his and Chapelon’s eminence, de Caso’s tongue was very much in evidence. When he had finished, there was silence and the chairman of the committee said: ‘Tell Monsieur Chapelon we don’t need to see him.’ Nevertheless, both Chapelon and de Caso were downgraded from their rank of ingénieur principal for six months, despite considerable protest from colleagues.