Economics · Industrials · Machinery
Why did diesel locomotives replace steam?
Within about a decade of 1948, the Class I railroads of the United States had replaced nearly all of their steam locomotives with diesels. Few industrial technologies have been displaced so completely, so fast - and the reasons, laid out in order, are a masterclass in how machines replace machines.
From the book The Railway LocomotiveChapters 57 to 62. Read the full book free as a PDF.
An engine looking for a transmission
Rudolf Diesel filed his patent in 1892 with deliberate intent: burn the fuel inside the cylinder, in air compressed so hard it ignites by its own heat, and approach the most efficient heat engine thermodynamics allows. The engine worked. Putting it on rails did not - because a diesel produces no torque at zero speed, cannot be slipped like a steam engine's piston, so it needs a transmission that lets the engine run while the locomotive starts. Diesel-mechanical and diesel-hydraulic drives exist, but the electric drive suited large American road and multiple-unit locomotives. The solution came from the electric streetcar: couple the diesel to a generator, feed the current to electric traction motors on the axles, and the electric drive becomes a clutchless, continuously variable transmission with full torque from a standstill. The diesel-electric locomotive was thus three inventions stacked - Diesel's engine, the dynamo, and Sprague's traction motor - and by 1913 experimental diesel-electric railcars were running in Sweden. The Soviet Union built a full-sized mainline diesel-electric, Gakkel's Shch-el-1, in 1924; America's first commercial diesel-electric, a 300-horsepower boxcab switcher, entered service on the Central Railroad of New Jersey in 1925.
Two proofs: glamour and tonnage
The diesel announced itself to the public with the Pioneer Zephyr: on 26 May 1934 the little stainless streamliner ran 1,015.4 miles from Denver to Chicago non-stop in 13 hours, 4 minutes and 58 seconds, averaging 76.61 miles an hour, with the nation listening by radio. It announced itself to the accountants with the FT: on 25 November 1939 Electro-Motive's four-unit demonstrator No. 103 left La Grange, Illinois, and spent eleven months hauling tonnage over twenty railroads and thirty-five states - 83,764 miles without a failure that mattered. Four units, 5,400 horsepower, under one driver. Trains magazine's editor David P. Morgan called it "the diesel that did it".
The Second World War both interrupted and accelerated the takeover. The War Production Board, rationing materials and capacity, restricted new diesel construction to freight locomotives and switchers - and in practice awarded the freight market to the builders already in it, Electro-Motive above all. Baldwin, the foremost steam builder of them all, was encouraged to keep building steam and not permitted to develop road diesels; by the time the war ended, Electro-Motive and ALCO had locked up the market between them, and the wartime FTs had run day and night in the hardest service the railroads could give them. Their availability figures - the share of their lives spent pulling rather than being repaired - were the numbers read aloud in boardrooms after 1945.

The five reasons
The first reason is efficiency. One comparison puts a typical steam locomotive of 1930 to 1960 at about six per cent of its fuel's heat turned into work, and a modern diesel-electric at almost thirty-six per cent at the drawbar. The second reason is availability and maintenance, which mattered in many services: a steam engine spent hours each day raising steam and weeks each year in washouts and boiler work - while a diesel started on a button and needed far less routine boiler work. Availability varied by service and railway, as a 1948 New York Central evaluation of its motive power showed, so the diesel advantage depended on the duty.
The third reason is infrastructure. Steam railroading carried a permanent army behind the locomotive: coaling towers and water tanks every thirty or fifty miles, pumping stations, water-softening plants, ash pits, roundhouses and the men who staffed them. The diesel needed a fuel tanker and a fitter. The fourth is the crew: diesel units plugged together under multiple-unit control and answered one driver's hand - a major labour saving. And the fifth is that the diesel was a product while steam had become a craft: Electro-Motive sold a standardised machine with interchangeable parts, published maintenance procedures and a field service organisation, built on an assembly line. The steam builders, masters of bespoke construction, discovered the market no longer wanted craft. It wanted units.
The holdout that proves the rule
The last stand was the rational one. The Norfolk and Western was a coal road sitting on the finest steam coal in America, served by its own superb Roanoke Shops, and it built the country's last new steam locomotive - a humble 0-8-0 switcher - in December 1953. It held out until the diesels' falling price, the rising cost of steam's ancillary army, and the collapse of the parts industry around it made even the best-managed steam fleet untenable. On 7 May 1960 the N&W shut down its last mainline steam engines, and the argument was over everywhere. What was lost was real: a craft culture of boilermakers, fitters, firemen and shedmen wound down with the fleet, and the preservation movement, from the Talyllyn's volunteers of 1951 to the restored Big Boy 4014 that draws crowds across America today, is the evidence of it. The railway, unsentimental, moved on to count its savings.
Quick answers
How much more efficient is a diesel than a steam locomotive?
About six times, on the International Steam comparison: it puts a typical steam locomotive of 1930 to 1960 at about 6 per cent of its fuel's heat turned into work, against almost 36 per cent at the drawbar for a modern diesel-electric.
How fast did American railroads dieselise?
Almost completely within about a decade on the Class I railroads, with mainline steam ending on the Norfolk and Western in May 1960.
Why couldn't a diesel engine just be geared to the wheels?
A diesel makes no torque at zero speed and stalls under load from rest, so it needs a transmission that lets the engine run while the locomotive starts. Diesel-mechanical and diesel-hydraulic transmissions exist, but the electric transmission - engine, generator, traction motors - suited large American road and multiple-unit locomotives, with full torque from a standstill.
Sources
- National Park Service, Steamtown Special History Study, chapter 5
- Trains magazine, "1939's big breakthrough: The FT"
- National Museum of Transportation, GM No. 103 "The Diesel That Did It"
- Wikipedia, "Soviet locomotive class Shch-el-1"
- internationalsteam.co.uk, "Steam vs. Diesel"
- craigtouchton.com, "Steam's Last Stand"
- Union Pacific, "Big Boy No. 4014"
- Engine Museum, "History: the dawn of the diesel era"
- New York Central, "A Practical Evaluation of Railroad Motive Power" (1948)
