Goffs was once a busy desert peak town, created largely by the Santa Fe as a stop to water its engines. During the Second World War, when the 4-8-4s were at their busiest, some ten thousand US troops were stationed there at any one time. Today, Goffs is almost a ghost town, with the wind blowing the surrounding sands into ever-changing and hypnotic patterns. At times it can seem silent. When I stood there a few years ago – I had been driving the length of the legendary Route 66, which has bypassed the town for some years – I tried my best to imagine the sight and sound of one of these great 4-8-4s thundering up from the parched plains below, with its air-conditioned trainloads of Hollywood stars and business moguls heading west, iced cocktails in hand. Few Westerns of the time, with their galloping wagon trains, were as thrilling as these thunderous, long-distance expresses.
Over slightly easier terrain, Allen mentions a postal train weighing 1,590 tons sprinting the 134.5 miles from Dalies to Gallup, New Mexico, with a ruling gradient of 1-in-220, in 124 minutes. The slowest speed maintained uphill was 59 mph, with 90 mph and over wherever possible. Throughout the 80 mile uphill section, the 4-8-4 produced about 5,500 ihp. The Santa Fe was to make the most of its steam locomotives until the last of them was withdrawn in 1959.
The quest for athleticism and muscle had become something of an obsession for American railroads and their engineers in the late 1930s, as they aimed to meet ever more demanding requirements from their operating departments. In the event, their timing was spot-on, because when the war came in 1941, the country was to be well served by locomotives like the Santa Fe Northerns. In 1938, the Association of American Railroads conducted tests to see what horsepower was needed for a 1,000 US ton (892 imperial tons) passenger train to run on the level at 100 mph. And it wanted to know if any existing locomotive type was equal to this demanding target. The one locomotive that got close was the Jabelmann-Alco FEF 4-8-4 815, which reached 102.5 mph on a 1-in-500 to 1-in-2,000 falling gradient. In the reverse direction, 815 reached 89 mph.
A design developed during the war which could have managed this with ease was Kiefer’s commanding Niagara class Northern for the New York Central Railroad. Designed jointly with and built by Alco, the Niagara was the biggest locomotive Kiefer could shoehorn into the New York Central’s relatively restricted loading gauge; to keep within this, the steam dome, normally prominent on the crest of a locomotive’s boiler, was replaced by a large internal steam-collector pipe. In his New York design office, Kiefer had drawn up plans for a prototype two-cylinder Niagara with interchangeable 6 ft 3 in and 6 ft 7 in driving wheels and a boiler pressure of either 275 or 290 psi, a four-cylinder 4-4-4-4 duplex, a 4-8-4 equipped with poppet rather than piston valves, and a standard production machine with 6 ft 7 in driving wheels, piston valves, and a 275 psi boiler. The duplex project was dropped early on, while tests were made on the other prototypes, which were soon joined by the first of the production engines.
There was not a great deal to choose between the variants. The production engines, class S1b, were thoroughly tested. They could produce a maximum 6,600 ihp at 85 mph and 5,050 dbhp at 62 mph, while their 8 ft 4 in diameter boilers generated steam at an impressive 110,000 lb per hour. Fitted with every possible labour-saving device, these 210 ton engines were as powerful and robust in everyday service as Kiefer and Alco had designed them to be. They ran heavy, fast, long-distance trains with easy competence and, with daily servicing reduced to two and a half hours, they were soon racking up 24,000 miles a month. The twenty-seven Niagaras were every bit as fast as, and exerted a much higher maximum power than, contemporary diesels, and even in terms of turnaround, maintenance, availability, and cost per mile, they ran the diesels that were to replace them within a decade a very close race indeed.
In June 1947, the Institution of Mechanical Engineers celebrated its centenary with events in London. These included a paper by Kiefer on the development, testing, and operation of rival forms of motive power on the New York Central Railroad. Back in New York, he had the paper published by the Steam Locomotive Research Institute as A Practical Evaluation of Railroad Motive Power. A copy of this small red book sits on my bookshelves between handsome illustrated tomes on American trains, locomotives, and railway history. It is a very prized volume indeed.
What Kiefer’s book shows is just how close he came to making a case for modern super-power steam at the very moment that the diesel was about to take over – indeed, just a year before the New York Central itself declared an end to steam development, due, primarily, to the unreliability of coal supplies during the prolonged miners’ strikes of 1946 and 1947. During 1946, six Niagaras were pitted against six 2,000 hp twin-unit E7 diesel-electric locomotives up and down the 928 mile Water Level route between Harmon, New York, and Chicago. The steam locomotives made 310 trips, the diesels 349, with fifteen coaches of air-conditioned stock weighing 1,005 US tons (897 imperial tons). Although the twin-unit diesels, which cost half as much again as a Niagara, could produce a maximum of 3,320 dbhp, compared to the 4-8-4s’ 5,050 dbhp, the steam engines had been designed with a large degree of excess power capacity over that required in ordinary service; this was so that they could work the most demanding trains economically and be able to make up time easily if a service was held up for any reason.
The results really are nail-biting stuff. The acceleration tests remain fascinating. A Niagara could accelerate its train along level track from 0 to 60 mph in 3.5 minutes (the test reports were published using decimals), covering 2.1 miles as it did so. A twin-unit diesel-electric needed 4.73 minutes to reach the same speed, over 3.2 miles. Where the twin-unit diesel was up to 80 mph in 14.17 minutes over 14.8 miles, the Niagara was there in 6.36 minutes and 5.1 miles. A triple-unit diesel-electric could pip the 4-8-4 to 60 mph, in 3.06 minutes over 2 miles; but it was slower than the steam locomotive to 80 mph, taking 6.51 minutes and 6.1 miles; and way behind the Niagara as it stormed up to 100 mph, when the steam locomotive took 16.5 minutes and 21.3 miles, against the diesel-electric’s 26.5 minutes and 37.8 miles. The Niagara was faster accelerating from delays, too, whether from 35 to 60 mph, 35 to 80 mph, or 60 to 80 mph – 2.86 minutes compared to the diesel’s 3.45 minutes in the last instance.
Kiefer himself was careful to present a balanced picture. ‘We have no sacred traditional standards, nor preconceived ideas or preferences with respect to the kinds of motive power used,’ he wrote, ‘but have striven to the limits of our collective abilities to provide units best suited to meet the changing necessities of transportation by rail in which numerous and varied problems are involved.’ He was clearly proud of the Niagaras, although he made it clear that the long-term future lay with electric traction. He addressed the question of relative thermal efficiency, an issue often used by its critics to belittle the steam locomotive. Where a diesel-electric offered an overall efficiency of 22 per cent – that is, 22 per cent of the fuel it burned was converted into energy used to move a train – a steam locomotive could manage just 6 per cent. This figure does sound low – it was later doubled by Chapelon and Porta (see Chapter 7), and there is still a promise of greater things to come – but the diesel-electric appeared to be the better bet in terms of fuel usage. Even then, as Kiefer stressed: ‘Although the overall thermal efficiency of the Diesel locomotive may be four to five times that of reciprocating steam, it should be recognized that without this advantage the cost of Diesel fuel oil would be prohibitive.’
With all the figures added up and computed, the total annual cost of the 4-8-4s was $1.22 per mile, and for the twin-unit diesels $1.11. Given that the price of diesel oil was to rise way above its 1946 level – and well above that of coal – the figures might have been in favour of the steam locomotive just a few years later. And if the New York Central Railroad had wanted to run even heavier or faster trains, then it would have been only fair to pit the Niagara against triple-unit diesel-electrics, producing 4,980 dhp, in which case the estimated cost per mile for the diesels would have risen to $1.48. Even then, three E7s
were no match for a Niagara in full cry above 60 mph, when its drawbar horsepower was higher. However, it must be remembered that the low thermal efficiency of a Niagara meant that it burned three times the number of heat units for the same power output, so it would have needed a 50 per cent increase in the cost of oil above the 1946 price to equalize fuel costs.
Impressive though their showing was, the performance and economy of the Niagaras could have been improved. One abiding characteristic of so many US steam locomotives was the high exhaust pressure of their blast-pipes, which was necessary to produce sufficient draught for their high-output boilers. This resulted in power loss caused by back-pressure on the pistons and, partly as a result of this, the figure Kiefer cited for a Niagara of 16 lb steam to produce 1 ihp was higher than both the LNER A4 and LMS Coronation Pacifics, and considerably above the 11.2 lb of Chapelon’s 5,500 ihp 242A1. If the 1,400 hp absorbed by draughting at maximum power in the Niagara could have been reduced to 500 hp by using a triple Kylchap exhaust – thereby reducing back-pressure – the locomotive could have produced a maximum of 7,500 ihp and used significantly less coal.
As it was, a series of annual coal strikes which began during the war led to large numbers of train cancellations in 1946 and again in 1947. The effect of these was so great that many railroads made the decision to drop their dependence on coal and to replace steam with diesel. The American railway press went so far as to describe the mining union’s leader, John L. Lewis, as having done more to undermine steam than any living man. And the diesel lobby was anyway very much in the ascendant, armed with new business techniques which were to eliminate main-line steam within little more than a decade.
On a happier note, what was a Niagara like in regular service? I cherish a photocopy of an article, ‘Riding a Niagara of the New York Central’, written by Dr W. A. Tuplin, chief engineer of David Brown, a company specializing in the manufacture of gears and tractors, which went on to make the famous DB series of Aston Martin sports cars from 1947. ‘“Let’s see it!” said the engineer from his seat ten feet above me after I had shouted that I had a permit to ride with him. After all, 7.30 on a Sunday morning is a strange time for a visitor, and the man in charge of the New York Central hauling the Mohawk from Buffalo to Albany was justifiably cautious about a proposed invasion of his eyrie. My letter from a Vice President immediately resolved his doubts.’
So up climbed the English engineer on to the spacious, almost room-like, footplate of Kiefer’s finest. The 900 ton Mohawk was an express train from Chicago to New York, with forty points of call, booked to average 41 mph overall. Between these many stops, however, the big engine accelerated rapidly and was worked up to the line limit again and again, with mile-a-minute runs readily achieved between stations set as little as fifteen miles apart. All this was done with an easy competence that made the working of the Mohawk seem little more than a Sunday stroll.
So many things surprised Tuplin, a man who had ridden on the footplates of many British locomotives. At stations, the giant fire-bed – the mechanically stoked 8 × 12.5 ft grate – appeared to be lifeless, flaming into life only as the engineer, sitting comfortably in his leather-upholstered seat, controls ready at hand, cracked open the regulator and the exhaust draught set it ablaze. The exhaust seemed mild and tinny, while ‘not even a turbine-driven engine could be expected to run any more smoothly than did the big two-cylinder engine, and no diesel locomotive is any easier to handle’. At 70 mph and above, the exhaust was all but inaudible, despite the fact that the Niagara was producing an average of 3,100 dbhp throughout the journey. Only once on his memorable ride did Tuplin hear the locomotive in any way extended, and that was as it climbed an 8 mile 1-in-132 gradient between the Mohawk river and the Hudson Valley. ‘Although the engine appeared to be working in an effortless manner,’ Tuplin enthused, ‘the start-to-stop times on this journey were what would be regarded as phenomenal in Great Britain with an engine hauling five times her own weight.’
Plagued at one point by cracks in their silicon-steel boilers – designed to save 3 tons in weight – the Niagaras, refitted with normal carbon-steel boilers, were taken out of service between May 1955 and July 1956. On its last passenger run, with train No. 416 from Indianapolis to Cincinnati, 6015 left 34 minutes late, making up 28 minutes on the way. Clearly, it still retained a surplus of power. It had replaced a failed diesel. Sadly, no Niagara was preserved.
The Pennsylvania Railroad took a very different engineering path from the rival New York Central Railroad as average speeds and train loads increased in the late 1930s with the ending of the recession that had begun with the Wall Street Crash in 1929. Remarkably, the Pennsy relied almost entirely on its 425-strong fleet of two-cylinder K4 Pacifics, built between 1914 and 1928, to work its heaviest, fastest, and most prestigious express trains into the 1940s. These were excellent engines, but by this time they had to be worked in pairs to keep the fastest and heaviest trains to time. The Pennsy’s main line from New York through Harrisburg, Pittsburgh, and Crestline to Chicago was sixty miles shorter than the New York Central’s but, running further south, it had to cross the Allegheny Mountains. The magnificent 220 degree horseshoe curve that took the line west of Altoona, the Pennsylvania railway town where the company built and tested many of its own locomotives, presented a formidable challenge to engines and their crews. Going west, it climbed 12 miles at 1-in-52, curving all the way to Gallitzin, with 23 miles at 1-in-87 in the other direction.
On a 1945 visit to the USA, the British locomotive engineer Stewart Cox rode a heavy train hauled by a pair of K4s over the 114 miles from Altoona to Pittsburgh. The steaming of the locomotives was excellent. ‘The other aspect,’ Cox reported, ‘was the deep carpet of smoke-box ash which flanked the line-side reaching up to a hundred feet into the pinewoods, and which was due to the terrific loss of unburned fuel arising from the very high combustion rates, accentuated by the pulverizing of fuel as it passed through the mechanical stokers. The load had to be got over the mountains, even if boiler efficiency was only 50 per cent, and the volcanic exhausts produced a symphony of sound the like of which we will never hear again.’
In 1937, F. W. Hankins, chief mechanical engineer of the Pennsy, formed a committee with S. B. Ennis, H. M. Glaezner of Baldwin, and W. E. Woodard of Lima to prepare designs for a locomotive to haul trains of 1,200 US tons (1,071 imperial tons) at 100 mph on level track and to develop 6,500 ihp at this speed. The result was 6100, the S1 class 6-4-4-6, built in 1938–9 in the Pennsy’s Altoona shops. This was the railroad’s first rigid-frame duplex locomotive with two sets of outside cylinders. A very large boiler with a 132 sq ft fire-grate made it necessary to use three-axle leading and trailing trucks. Despite its excellent performance, 6100’s extreme length – 140 feet – restricted its route availability and the Pennsy asked the locomotive builders for a more compact duplex locomotive to haul 880 tons.
The result was the fifty-two T1 class 4-4-4-4s, a class of very fast, streamlined, four-cylinder duplex locomotives. Two prototypes were built by Baldwin in 1942 and the production batch in 1945–6 by Baldwin and the Altoona works. Sadly, the production engines lasted just two years in charge of crack Pennsy expresses, including the Broadway Limited from New York to Chicago, and were taken out of service altogether within five or six years of the last engine being delivered in August 1946.
The T1s were remarkable machines. The idea had been to shape an engine with smaller cylinders than the classic two-cylinder American 4-8-4 and so halve the piston thrusts, reduing piston speed and the ‘hammer blow’ that the heavy driving mechanism of large two-cylinder machines transmitted to the track. Clad in shark-nosed streamlining designed by Raymond Loewy, the first two locomotives – 6110 and 6111, the latter fitted with a Franklin booster on its trailing truck – were ready for testing in spring 1942. Visually, they were particularly striking machines, looking menacingly fast even when standing still. In practice, the streamlining encouraged exhaust steam and soot to drift over the boil
er and into the cab; when the new engines were painted the Pennsy’s almost malachite take on Brunswick green, they appeared to blacken after only a few days in service. Their specification was equally impressive. Hugely long, at 122 feet, which was 10 feet more than a Niagara, the engines sported 200 ton tenders mounted on two eight-wheeled trucks. Boilers were pressed to 300 psi, tractive effort was 64,650 lb, and, on test at Altoona, 6110 developed a peak of 6,666 ihp at 100 mph with 20 per cent cut-off, with a boiler evaporation rate of 105,500 lb per hour and a pressure drop of only 7 per cent between boiler and cylinders. The minimum specific steam consumption, at 5,000 ihp and 96 mph, was 13.6 lb per ihp per hour, the lowest recorded in forty years of testing at Altoona. On road tests, 5399 attained 94 mph on level track with a train of 1,178 US tons.
Performance, especially over the lightly inclined 280 mile section from Crestline to Chicago, could be sensational. On test, 6111 ran 69 miles at the head of a sixteen-car, 912 ton train at 102 mph, while with an almost identical load, 6110 is said to have run half of the 280 miles of the same route at a mean speed of 100 mph, with steam flowing freely through its large poppet valves. There were rumours throughout the brief lives of these charismatic machines of speeds of up to 125 mph. These should not be dismissed. The T1s were undoubtedly among the very fastest and most powerful of all passenger steam locomotives. Sadly, though, there are few details to back up such claims. But, although the T1s could fly, they – both the prototypes and the fifty production series that went into service between November 1945 and August 1946 – suffered from not one but several Achilles heels.
The first was slipping. Although careful driving could alleviate this problem to an extent, the two uncoupled sets of four driving wheels could easily slip differentially and at varying speeds. This could make hill-climbing tricky and acceleration irregular, while, from a maintenance aspect, the use of Franklin type A oscillating-cam poppet valves, with the driving gearboxes inaccessibly located, proved a problem. Johnson at Baldwin would have preferred rotary-cam poppet valves, with an easily accessible external Cardan shaft drive, similar to that used on the British Railways three-cylinder Pacific 71000 Duke of Gloucester.
Giants of Steam Page 24