Home » The UAC TurboTrain Was America’s Failed Plane-Engined High-Speed Train Of The Future

The UAC TurboTrain Was America’s Failed Plane-Engined High-Speed Train Of The Future


The 1960s were a wild time to be a traveler. Jet aviation made the world feel a bit smaller by making far off lands accessible in less time. But on the ground, people still got around in slow cars and trains. Japan’s Shinkansen proved that rail service could be made fast, and America wanted in. One answer to the call of high-speed rail was the UAC TurboTrain. This passenger train tilted in curves and was powered by gas turbines meant for aircraft. Yet, just seven were built and the trains were in service for barely over a decade.

In 1964, Japan showed that in a world of jet travel, trains could still be relevant. The first-generation of Shinkansen trains operated on the Tōkaidō high-speed line going as fast as 130 mph. The “bullet train” caught the attention of not just railfans and commuters, but of companies and governments.

As the U.S. Department of Transportation’s Federal Railroad Administration notes, the government began looking into high-speed rail right around the launch of the Shinkansen. A year later, President Lyndon B. Johnson signed the High Speed Ground Transportation Act of 1965 into law. In the signing of the Act, President Johnson remarked:

In recent decades, we have achieved technological miracles in our transportation. But there is one great exception.

We have airplanes which fly three times faster than sound. We have television cameras that are orbiting Mars. But we have the same tired and inadequate mass transportation between our towns and cities that we had 30 years ago.

The Act initially authorized $90 million for demonstrations and studies of high-speed ground transportation. The Northeast Corridor Demonstration Project. As reported by the Chicago Tribune in 1967, the project was meant to show that a high-speed train could convince passengers to take a train to commute between Washington D.C. and New York City rather than drive or fly. This would reduce crowding on highways and in airports. A win for everyone.

The project was in two parts. One part would become the Budd Metroliner, an electric train built for high-speed. The other part would take a different route.

Marty Bernard

The United Aircraft Corporation entered the Northeast Corridor Demonstration Project with what it called the TurboTrain. This train would not be powered by electricity, but by a gas turbine. And it could achieve faster speeds than a Shinkansen thanks to an aerodynamic shape and technology that allowed the train to tilt through curves.

Design of the TurboTrain was initially assigned to the Corporate Systems Center of UAC before being passed to Sikorsky Aircraft, another UAC division. An archived brochure for the train details not only the technology involved, but what engineers had to overcome. Japan’s Shinkansen (and later, France’s TGV) run on tracks specifically for high-speed rail applications. They enjoy long, swooping curves, no grade crossings, no freight traffic, and trains go through obstacles rather than around them.

Konica Minolta Digital Camera

Unfortunately, trains in the Northeast Corridor Demonstration Project had to work with older infrastructure. But as Sikorsky noted in the brochure, a typical train under the influence of centrifugal force in a curve leans outward. That’s fine, normally, but too fast and you risk both comfort and safety.

Since the TurboTrain would be all about speed, UAC decided that it would tilt into the curves. To do this, UAC purchased the patents to a Chesapeake & Ohio Railway study. C&O was developing an articulated train in the 1950s.

Us 3424105 Aa
United States Patent and Trademark Office

These train cars shared a common bogie (truck) instead of each car having two bogies each. The cars sat on A-arms and the forces in a curve would cause the cars to naturally lean into the curve. Air springs were there to smooth out the leaning motions.  The body was made of aluminum.

The dome that the UAC trains would become known for was also a C&O creation.

Us 2859705 A
United States Patent and Trademark Office

Sikorsky noted that by allowing the forces to be down low, the TurboTrain was able to round curves 30 to 40 percent faster than a conventional train. The company compared the experience to that of when an airplane banks in a turn.

The trains had a distinctive look, too. On the ends of all TurboTrains were rounded power cars. Power cars housed passengers, the crew, and the gas turbines. The crew sat up high and passengers riding in the power cars were able to view what the crew was doing through glass windows.

John Stewart

Common bogies between the cars meant that each trainset was semi-permanent. Since the cars were essentially mated to one another it wasn’t a quick task to swap cars like in a conventional train. Thus, each trainset had power cars on each end. When the train needed to change directions, the crew would just move to the other power car.

This promotional video further explains how the suspension works:

And that power came from an interesting powerplant. Power came from Pratt & Whitney Canada (then a UAC subsidiary) ST6 turboshaft engines. It’s a development of the PT6, a turboprop engine found in numerous aircraft with examples like the Cessna 208 Caravan and the Beechcraft King Air.

Whereas these engines would drive a propeller on a plane, here they’re driving a gearbox, which drives the train’s wheels.

Turbotraindiagrams 0004

In this application, each ST6 is fed from diesel fuel and is rated at 400 horsepower. The power cars could hold multiple engines. One of the engines of a consist provided head-end power, or electrical power to the train and passenger cars. Sikorsky claimed that a TurboTrain with seven cars and 2,000 horsepower (five ST6 engines) would carry enough passengers to take 150 cars off of the road.

Weirdly, the company keeps with the horsepower figure and says that the train would take 30,000 horsepower off of the road, or about 200 horses per car. It’s unclear how Sikorsky arrived at that figure.

Either way, these trains were fast. On December 20, 1967 a TurboTrain reached 170.8 mph during acceptance testing on a high-speed test track on Penn Central’s mainline. UAC’s creation not only beat the competing Metroliner project, but blasted past the speeds of what the Shinkansen could do back then.

The TurboTrain was put into service in both the United States and Canada in 1968. Units for Canada were built by the Montreal Locomotive Works and initially run by Canadian National Railway. In the States, they were built at the Pullman Works in Chicago and initially operated by the New Haven Railroad. Check out this video on the train!

There are more of these wild videos, like this one that shows the TurboTrain taking off like a spaceship:

As Spacing Magazine writes, riders were treated to a luxurious experience not unlike that of airliner concepts of the day. Riders sat in recliners, walked on carpet, and enjoyed an atmosphere with soft lighting and even draperies. The trains even featured meal service. Seatbacks had folding tables and travelers would put their luggage into an overhead compartment, too.

Unfortunately, these trains would not see those speeds in service. instead, as Canadian news site the Walrus wrote, they would instead travel at a more leisurely 95 mph. And speeds would really only be the start of the train’s issues. In fact, a TurboTrain was involved in an accident only an hour into Canadian service.

As the Walrus reported, the driver of a meat truck tried to beat the TurboTrain across a grade crossing. They failed, and the train reportedly sliced through the truck like butter. Thankfully, the driver survived, but the crash highlighted a difference between American high-speed rail and the Japanese system that inspired it. As Spacing Magazine reported, that crash and the existence of 300 grade crossings meant that those slower speeds were actually mandated.

The Walrus explained further problems. These trains were run on the same rails used by other trains, so they had to slow for tight curves and they still had to worry about car drivers trying to beat the crossings. As I said before, Japan got around this issue by making sure that its Shinkansen didn’t have grade crossings and that the trains enjoyed long, gentle curves.

The site also noted that at the speeds achieved during testing, a TurboTrain could get from Toronto to Montreal would have taken two hours. That–after factoring in the entire process of getting to and from an airport to fly–made it faster than flying and way faster than driving. However, at the actual speeds that the train went, it took four hours, or only an hour faster than a car.


But it got even worse, from Spacing Magazine:

The brakes seized in winter and the exhaust from the engines in the forward locomotive spat soot over the windows. A Turbo caught fire in Toronto in 1970 and frequent technical glitches triggered several prolonged hiatuses, the longest of which took the trains out of service for several years.

It would reportedly take until 1974 for Canadian National and UAC to iron out these issues. By then, TurboTrain operations in the States moved to Penn Central Railroad then to Amtrak. The issues weren’t limited to Canada, either. In 1976, Amtrak sidelined its two prototype trainsets and one trainset that it bought from Canadian National. The reason cited for this? As the Eugene Register-Guard newspaper wrote, an Amtrak spokesman called the trains maintenance nightmares. And with a next generation of train cars on the way, Amtrak decided enough was enough.

Back in Canada, the remaining trainsets continued to run, but with better reliability. As Toronto Star news reported, the trains were on time 97 percent of the time and had a 98.6 percent availability.

Carl H. Sturner

Still, the trains were hampered by more roadblocks. As Spacing Magazine notes, not only did the trains have to slow down for grade crossings but freight trains also took priority on the lines. That only further erased the proposed time savings from the TurboTrain. Via Rail took over operations from Canadian National in 1978 and ran the trains until 1982. Then, even Canada called it quits. TurboTrains were replaced with conventional diesel-powered trains.

The TurboTrain isn’t the only application of turbines for a train. You can find turbine power in trainsets throughout history, and even Amtrak had another, slower turbine train with the Turboliner. You may hear more about these in the future!

Today, you won’t find a UAC TurboTrain anywhere. Just seven trainsets were built and all met the scrapper. They now only exist in riders’ memories, the internet, and scale models.

Share on facebook
Share on whatsapp
Share on twitter
Share on linkedin
Share on reddit

34 Responses

  1. Many of the features were reused in the TGV001 ( the TGV prototype ) that was built around the same time.

    But the 1973 petrol shock killed the gas turbine powered TGV and it moved to electricity for power source.
    The Japanese philosophy was reused in France for the High Speed track sections : no level crossing, but there’s a huge caveat. In Japan the standard gauge is uncommon at best outside the Shinkansen network, so there’s no interoperability in tracks, while in France ( and in western Europe ) High Speed trains can use old regular tracks for part of the travel. ( like the last few kms to reach the city center railway station )
    While it was not used in France, the Italians developped ( and use, along with the English ) the pendulation system. ( Pendolino )

  2. Mercedes. Come on. You’re not going to mention the American version put into service by Union Pacific? Because holy shit, buckle your seatbelts, folks. Shit’s about to get WEIRD.

    The Canadian TurboTrains only ever ran on jet fuel. Yeah, literal jet fuel. Cost a fortune but worked. But ALCO-GE said “yeah, that sounds expensive. How about… an oil fired gas turbine?” And thus the ALCO-GE GTEL was born in 1948.
    But no, no, this is not the insane part. It entered revenue service as the Union Pacific GTEL which was wildly different from the TurboTrain. Same general concept, many of the same people involved, but it was for revenue freight service. So… they used “Bunker C” oil, because it was cheap. And sure it coked everything all to hell and fouled every surface, but it was c-oh. It’s no longer cheap.
    So in 1953 they converted it to run on compressed propane because THAT was cheap. And hey! It worked! No more fouling! There was the whole ‘explody boy’ problem of hauling 40,000 gallons of highly compressed propane directly behind something spitting out exhaust gases that could exceed 1500C. But small problems.
    So they built a second generation. 14 more locomotives introduced in 1954. They did MU control tests in 1958, and worked so great that, uh, they were built to run exclusively with diesel locomotives assisting. Oops.
    No problem. Technology is hard. Enter the third generation where things get TRULY fucking bonkers. Not many steam locomotives left on the rails in 1955. So UP ordered an 8,500HP GTEL “X-1.” Which is still the most powerful single prime mover ever built. But man, those fuel costs. What fuel was cheap?
    Yeah. Here’s the bonkers. In 1962 they took an ALCO PA-1 cab, a GN W-1 chassis, and a modified turbine unit pulled from a second or third gen locomotive. Then they hooked a coal tender from Challenger 3990 up to the back of it because, yeah folks. “Holy shit, they didn’t.” Oh, no, not that.
    They made it run gasified coal, I shit you not. Not coal-to-steam. Actual gasified coal. As you can probably guess, shooting tiny particles of highly abrasive coal through precision machined lightweight turbine blades was … not great for longevity. Hey now instead of sludge and fouling, you’ve got coal dust and half the blades are so eaten away they double as musical instruments! 20 months in, and they scrapped it and pretended it never happened.

    1. This isn’t an exhaustive history of all turbine-powered trains, just one of them that was a massive failure. Clearly, people like trains, so I have to write about more of them!

      Or just read your excellent short essay, there. 🙂

    2. I’m guessing she didn’t bring up the UP GTELs either because she is planning to write about them separately (in the second-to-last paragraph she says we may hear more about turbine trains in the future) or because they’re likely one of the best-known or possibly the best-known turbine locomotives in the US. If you bring up turbine locomotives to any railfans I know, UP’s GTELs are the first thing they will think of.

    3. That is some great trivia about the gasified coal! I was at least vaguely familiar with UP turbine-powered GTELs, but had no idea they ever ran that stuff through one of them. Reminded me of GM’s balls-out nuttiness of running a ’78 Eldorado on coal. Oh, the future that could’ve been!

  3. I almost got to ride a CN Turbo Train in 1975 but true to form it was broken and we rode a regular CN service from Montreal to Ottawa.
    The Oregon Zoo’s narrow gauge line has a Train made to look like the 1950s GM gas turbine Train although it’s a diesel

  4. “Common bogies between the cars…”

    Common single-axle bogies, no less, as you can see from about 6:00 in the first video. They were yet another problem, giving poor low-speed ride and “hunting” around curves, besides the difficult maintenance.

  5. In the US, the issue of track ownership is huge: the freight-hauling railroads own the tracks, and those heavy trains make maintaining tracks consistent with a decent ride for high-speed passenger trains impracticably costly. And that’s before even considering all the grade-level crossings.

    I got to ride the Shinkansen between Kyoto and Osaka several times decades back. Coolest thing ever for an 8yo.

  6. Living on the Toronto-Montreal route I would see these once in a while.
    Thanks for the tech breakdown I just figured they were turbine-electric and was wrong.
    Here the fight between CN freight vs. VIA passenger priority still exists with the passenger trains getting delayed.

    1. Amtrak-owned lines running Philly and New York commuter rail are great fun for this reason. Nothing better than sitting on the platform waiting for a behind schedule amtrak train to pass. The conductors can get pretty salty over the intercom.

      1. My experience was the opposite. My family sat on the Amtrak, waiting for behind-schedule freight trains to get out of the way. (This was the Texas to Chicago line.) We stopped multiple times, because the freight trains had priority.

  7. “Common bogies between the cars meant that each trainset was semi-permanent. Since the cars were essentially mated to one another it wasn’t a quick task to swap cars like in a conventional train. Thus, each trainset had power cars on each end. When the train needed to change directions, the crew would just move to the other power car.”

    Not to nit-pick here, but that’s kinda been commonplace for quite some time on commuter railroads and subways running on electric or pantograph. I could go on for quite a bit about this subject…

    Otherwise, fascinating read about a tech that I had only heard about in yearly re-cert courses (which obv never really cover the topic in-depth), because everyone just wanted the signature and an early quit for the day. ha.

    More train stuff, please! 🙂

  8. Now Mercedes is writing about trains??! Sweet!

    Fun fact, in the Rush song Red Barchetta (a great song about a car!) the lyrics reference a Turbo crossing the borderline. They were probably inspired by the CN version of this in Canada at the time.

    1. I saw in a few previous articles that readers want trains, too. And, truth be told, one of my favorite stories to write about at the other place was that abandoned train in Tennessee. So more trains you get!

  9. This was a great read! I had seen a picture or two of these over the years, but never studied up on the details. Would have been great if they had been the stepping stone to our own Shinkansen here in the States. Or, really anything faster, less-wobbly and more on-time that our current coast-to-coast Amtrak cruisers (as much as love riding on the California Zephyr).

      1. New York Central did just that in the 60s. The took a Budd RDC, stuck a few?jet engines at one and a?streamlined “shovel nose” at the other end; tested it a decided it wasn’t useful. FWIW the NYC system which became Conrail and Metro North was extensively electrified so electric multiple units made more sense.

  10. Yup… these trains were a debacle. What’s the point of a faster train if you don’t have the infrastructure to support the higher speed? It’s just a waste of time and money.

    Instead of spending money on the trains, they would have been better off enhancing the track infrastructure so there would be fewer holdups due to freight traffic as well as upgrading stretches of track to enable higher speeds.

    As it stands, the infrastructure is such that even conventional diesel-electric locomotives being used can’t run at full speed due to substandard track infrastructure.

    1. Absolutely. As with many of our problems with infrastructure, and, indeed, broader social issues, the problem is primarily one of policy and technology only secondarily. This is true of transportation, just as it is true for housing, health care, and most any crisis you care to name in the U.S. today. Technology must follow from and serve bold and thoughtful policy. In the U.S. we think we can let the status quo go undisturbed through techno-fixes. We think that we can make housing affordable by engineering mass produced modular housing while leaving untouched the commodification of land and the unfettered use rights of individual owners. We think we can fix transportation by throwing subsidies at car companies or letting silicon valley last-mile vampires flood cities with scooters funded by venture capitalism instead of giving up on the sunk cost of auto-centric infrastructure and investing in publicly funded transportation. These trains were a technical marvel, but they were not matched with a public commitment to make them work, a policy scheme that recognized that policy decision are almost always, in some way, zero sum–something must give to make something else, something new, work.

  11. The current Acela trainsets on the Northeast Corridor tilt as well, but you’ll only really notice on the few curves where they can stretch their legs in MA and RI. The problem in the US is track ownership, and the absolute panoply of state and local governments (not to mention Fed) who would need to agree on something because it makes sense, and only because of that. I’m a frequent NY to Boston passenger and I’d say it’s not bad, comparable to the SNCF in comfort, but there’s only a single line, and it doesn’t go to the Cote D’Azur

    1. IIRC there’s a new ACELA coming that’s basically an Americanized TGV, but due to the tracks it’s never going to reach the speed his European Cousin can reach.

  12. Great article! Mercedes, your work title should be Secretary of Transportation. I’m learning things and having fun. Being naive, I’ve often thought that high speed railways would be ideal for running between the Midwest states and the West Coast.

Leave a Reply