The 'Car Train' - would it work?

There’s been a few threads lately regarding trains of various types, and why the U.S. doesn’t rely on them more. The answer is that the U.S. is much more distributed than Europe, the cities are bigger and more spaced out, and communities are spread out like dots in all directions. Building a feasible passenger train network between them all in a way that was convenient and not hellishly expensive seems like an impossible task. But some people think it’s important enough that they advocate rebuilding our cities and packing people in tighter to facilitate train movement.

So this got me thinking about trains in general, and what they’re made of and what makes them so efficient. These are the beneficial engineering aspects of trains:

  • they have a diesel-electric locomotive, which is very efficient.
  • They roll on steel tracks using steel wheels, which keeps rolling resistance down.
  • They are very long, which minimizes aerodynamic drag. The frontal section of a train is miniscule compared to its overall size.

What makes a car so inefficient?

  • The engine has to be sized for decent acceleration, meaning it bigger than it needs to be for cruising at highway speeds.
  • The tires are rubber for good traction, which increases rolling resistance.
  • The car is small on comparison to its frontal area, making it aerodynamically inefficient.

Now, think about the new generation of cars that are coming - plug-in hybrids and pure electrics. A plug-in hybrid with a diesel generator is basically a small diesel-electric locomotive.

One of the problems with electric cars is the size of the battery needed to travel long distances. Plug-in hybrids solve this by adding a generator.

It occurs to me that if you take five or ten electric or hybrid vehicles and put them together, you’ve got a mini train.

The car train.

Here’s how it would work: You’re planning to travel to another city. You go on the internet, and put in your departure time. Other people do the same. When enough people are departing at the same time and want to participate in a train, they are all issued a code number that allows them to identify themselves. You drive to the departure area, and a device in your car identifies all the cars that want to take part in the train. You all join up, connect cables from one vehicle to the next, and depart.

Computers aboard each car maintain separation and manage power flow to each motor. Some motors shut down completely. You only use as much motor as you need. It’s like the ultimate in variable displacement technology.

Each vehicle is right up against the next (we might even use trailer-hitch type connectors to lock the cars together). Therefore, you get the aerodynamic benefits of a real train.

Only the lead driver has to ‘drive’. Every other car responds to steering inputs from the lead car and follows it. Everyone else can sleep, or do work, or whatever.

As the train moves, computers calculate the drag of each vehicle and figure out its contribution to the energy use of the entire train. At the destination, a charge is assessed to each vehicle, the train stops at the ‘train station’ (little more than a roadside turnout), the train breaks up, and everyone continues on their merry way.

If you don’t like the purely ad-hoc mechanism, you could do this:

The ‘Commercial’ car train:

The lead vehicle is actually a commercial ‘locomotive’, which looks like a minivan and basically is full of batteries. The driver is a licensed ‘train operator’. They leave on a schedule. You just drive to a station, hook up to the next departing ‘locomotive’, wait for a few more to join up, and take off. The locomotive provides battery power to all the cars, and computers decide which car’s motors should be run. You get to your destination with your battery still full, and pay the locomotive your share of the energy cost.

It seems to me that if such a thing existed, you could use market forces to really drive efficiency. For example, you could buy a car with retractable ‘train wheels’, which would have extremely low rolling resistence. Once you’re in the train, you deploy your train wheels, and save money. Or perhaps the train’s computers would look for the most efficient engines and use those. If your car is picked, you get a discount to account for the fact that you’ve improved the entire train’s efficiency. So there’s a big incentive to buy highly efficient cars.

Any thoughts on this? The obvious flaw is safety - with vehicles packed so closely together, an accident multiplies in deadliness. A poorly maintained train car could break in a way that endangers the entire train. A lousy driver at the front endangers everyone else. So we’d need to solve that issue. But that seems like a much more tractable problem that cross-crossing the country with high-speed rail lines and solving the problem of how to get where you’re going once you’re off a regular train.

The beauty of all this is that it could be done with no infrastructure changes whatsoever. You could have ‘train-able’ cars, and once which aren’t, and they can coexist. You have all the benefits of a car at your destination, and most of the benefits of a train while traveling between cities. A ten-car train wouldn’t be much longer than a double-trailer semi.

Right now, I can imagine that this would make a good science fiction story, but how practical do you think it would be? How much energy would a ‘car train’ save, and would it be worth it? I’ve never seen this idea written up before that I can think of, and I’m wondering why.

In terms of cost and wear-and-tear on the vehicle you own, I don’t see a benefit. One car in the train breaks down or incurs a flat tire and everyone in the “train” is held up by it. And every person in the “train” had better have his bladder and bowels in synch with everyone else. Taking a commercial airline and renting a car at the other end has every advantage.

The benefit is in energy cost to travel. And wear and tear would play a part as well - If there are 10 cars in the train, your motor would be used less than 10% of the time, if at all.

Clearly, we’re not talking about today’s cars doing this. I’m talking about the next evolution of automobile, building on the fact that cars are then electric, or hybrid electric, and highly computerized.

As for the bathroom break, since only the lead needs to drive I can imagine such vehicles evolving to allow you to rotate the seats around into a club configuration for comfort, and maybe even have a small potty in them if need be. Or perhaps you’d simply signal the train that you want ‘out’, and it would stop at the next ‘train stop’ (a roadside turnout like a rest station), and you could decouple for a small additional fee to make up for the time and energy to get down from and back up to speed.

Are you talking about running these car trains on existing highways and city streets? Length would be a limitation – let’s say about 5 cars max in a conga line. That eliminates a lot of the economies of scale you’re proposing.

Another problem I see is that every car would need to be equipped with some sort of interface that allows it to be connected to the “locomotive” as well as every other car. Regular railroad cars aren’t designed to be operated independently, so they’re basically just big dumb trailers. But every automobile has its own engine, onboard computers, brakes, fuel system, etc. and they would all need to be synchronized.

Amtrak operates an auto train than runs from New York to Miami. You ride in the train and your car travels in a special car on the same train. You pick it up at your destination.

Why separation? Add interlocking bumpers & I think it’s a great idea.

Comparing your car-train to a passenger train is really the wrong way to go about this. High-speed passenger trains in Europe and Japan (which posters in other threads have cited as a model for us to follow) run on electric current from an overhead wire, not diesel-electric. Plus, a train is designed to work as a unit. The aerodynamics come from one car flowing smoothly into the next. Your car-train is made up of pieces designed to work well separately.

To justify linking cars together, it has to be more efficient than driving them separately; and enough of an improvement to be worth the effort to make it happen. Off the top of my head:

  1. The cars are close together. This is a benefit akin to drafting in racing, but it’s far from a free ride. And you could get this without linking the cars together. With all the computer control you’re proposing, just establish standards for wireless communications between cars. When one needs to slow down, it signals any car following closely to apply the brakes faster than a human could react.

  2. Linking the powertrains. I’m not sure there’s any benefit here. You say “If there are 10 cars in the train, your motor would be used less than 10% of the time”, but during that 10% of the time, it has to produce 10 times its usual output (minus the benefit you get by drafting). A hybrid car just doesn’t have that much power to spare to drag around 10 times its standalone weight. And if you designed the cars to do it, you’re right back where you started before hybrids.

  3. The extra weight to make this work. Just as an example, you suggested retractable train wheels for each car. That’s hundreds of pounds you’re hauling around during all the time that you’re not in a train, and that would eat up any benefits from the times you were in a train.

I did mention that you might want to physically connect the vehicles like railroad cars are physically connected, through hitches.

A ‘turnpike double’ tractor-trailer setup can be about 130 feet long. (two 48-53 ft trailers, plus the length of the tractor). That’s more like ten typical mid-size cars, or 18 smart cars.

True enough. But passenger trains also have much higher weights per passenger than cars do.

Yes, I mentioned ‘soft linking’ in the OP, where computers would be used to space them. The disadvantage I see is that al the motors then have to be running and constantly adding/removing energy to keep everything in line. Hard attachments might be better.

As for the aerodynamics, that’s a tricky one, and you’d have to do some testing. trains aren’t faired all that well together either. Assuming cars are designed from the start to be ‘train-able’, they’d probably have blunter noses and other features that allow them to fit well together. Think Scion xB designs - stubby, boxy cars. You might even carry some simply fairings in the trunk that you could attach to make the train more aerodynamic.

The advantage is that you’re always operating the engines right at their peak efficiency, and you’re minimizing driveline losses (the majority of wheels are free rolling, and not subjecting the train to transmission losses).

The other advantage, with the ‘commercial train’ concept, is that you’re only carrying around the batteries you need for city commuting, because the ‘locomotive’ contains the cross-country battery capability (or battery/generator, if need be). If you want to move frmo one city to the next, you connect to a big moving battery to supply the juice. So in-city performance is much improved.

The ‘train wheels’ were just a thought. Maybe a better solution would be tires that could be inflated to a much higher pressure when in ‘train mode’. Semi-truck tires have 30% less rolling resistance than car tires, in part because they are inflated to higher PSI levels. We don’t do that in cars for ride/comfort/handling, but you might be able to work out a better compromise if you’re in a car train.

Come to think of it - a car train might also be a good application for hydrogen fuel cells. The drawbacks of fuel cells include the need to build a hydrogen filling infrastructure, plus the cost of the fuel cell. But if the ‘locomotive’ were actually using fuel cells, then the locomotive company could store its own hydrogen for refilling, and you’d only need one fuel cell vehicle to provide electricity to all the conventional gas/electric hybrids in the train. Instead of needing hundreds of hydrogen filling stations in a city, you’d only need a handful on the outskirts to power the locomotives.

I love the concept but it sounds a little complicated. I don’t mean I don’t get it, but I don’t really see Joe and Mary Roadtrip getting on board right away. They would probably feel a little silly, and question how much money they were saving, especially when you factor in waiting for the train and everything else. What if they figured out they needed to be in a 7-car train to save money, so they drive 30 miles out of the way to the train station, wait a half hour, and then 2 people don’t show up? I bet they won’t try that again anytime soon.

The bathroom and length limitations already brought up are legitimate concerns, and there are others. No sightseeing. No pulling over to get a picture. No seeing a restaurant or Dairy Queen and spontaneously deciding you want to eat there.

I’d do it, by the way, just because it sounds cool. There are a lot of little issues, though.

Sounds something like this . Its a car that attaches to a monorail. Seems wildly impractical, just too much infrastructure needed. But its thinking in the right direction.

I think one way that trains have better fuel efficiency is that they only travel “highway” miles: they’re on designated routes which allow them to keep moving at a steady rate. A lot of car travel runs on open roads where the vehicles have to regularly stop and start up again, which uses up fuel at a much higher rate. I assume the car trains the OP described would only run on expressways, and if so, would they be that much more efficient than individual vehicles?

Wel, that’s the question. From what I can see, the efficiency gains would be from the added streamlining, the more efficient use of the drivetrain, and perhaps a decrease in rolling resistance if we could figure out how to have the vehicles work in a ‘train mode’.

Other advantages would be that no one other than the driver in front needs to pay attention to driving, so you could work, sleep, or chat while moving between cities, and you could size the batteries for electric cars more like plug-in hybrids, which would eliminate a lot of dead weight for vehicles that mostly city commute but still need to be able to travel between cities on occasion.

I’m not sure whether this breaks down into as much advantage as the logistical hassles of forming the train would cost, but it’s worth thinking about - especially compared to the alternative of building more passenger trains and track.

How would it handle? Would turning those two little wheels in the front overcome all the momentum behind them, or would they just begin skidding?

How would it tolerate crosswinds? I’ve driven across New Mexico twice and there are some brutal winds on that stretch of I-40. The cartrain would present a large broadside.

How would it handle braking? Imagine decelerating that behemoth coming down a mountain.

That’s why I said that the cars would all be connected to an information bus. That way, they could all actively brake, steer, and drive.

But I may be victim to hype of the whole train concept. I’ve heard so many people talk about how important it is to convert to trains and get rid of cars, supposedly to stop global warming and reduce energy needs, that I never bothered to check if it’s even true.

Luckily, the Department of Energy has a handy reference guide which show the energy cost of various transportation alternatives. And it turns out that passenger trains kind of suck. And in fact, buses and other forms of mass transit aren’t much better. This was surprising to me.

For example, from tables 2.13 and 2.14, we discover the following:

2006 Energy use of transportation, in BTU/passenger mile (car based on average of 1.6 passengers):

Transit Bus: 4,235
Automobile: 3,512
Train (Amtrack): 2,650
Train (Light Rail Transit): 2,784

So it turns out that the car fleet as a whole is already more energy efficient than transit buses, and not that far behind light rail transit and inter-city train service. And since that’s the auto fleet as a whole, I imagine that the most efficient cars on the road are already more energy efficient than trains. And in fact, since cars are more efficient on the highway than in the city, and the stats are aggregate, it may be the case that cars are already more energy efficient than trains between cities.

That’s something I never hear from train advocates.

So now the only advantages I can see to public transit are easing of congestion and the use of electric power for those that do. Congestion isn’t a big issue between cities.

The ‘car train’ might still be feasible because it might improve the efficiency of an already-efficient transportation mode, but it’s certainly not necessary to compete with trains.

Did you consider that your numbers might reflect the fact that we are adapted to cars being our primary mode of transportation?

You also seem to be glossing over that trains are not insignificantly more efficient. They look more like 25% more efficient to me. That’s nothing to scoff at when you’re moving 300 million people around the country. I think this is the kind of scale on which 1 or 2% would be considered huge.

Again, the car fleet number is based on the averages, which as I recall is about 22 mpg, and 1.6 passengers per mile traveled. Drive a 50 mpg car, and you’re doing significantly better than the train. Put 2 or more people in your car, and you’re doing much, much better.

Even a single person in a typical car will be doing about as well as public transit buses, it looks like.

So we shouldn’t mix arguments. If you advocate public transit, it would be to lower congestion and maybe to lower pollution in the city, but not because you want to lower how much energy we use to travel. That’s the way it looks to me.

We shouldn’t mix arguments? But the thread is about trains and now you’re talking about buses. And the averages are averages for a reason. The train still wins by a long shot.

The thread is about ‘car trains’. I thought you might need to do something like that to approach the efficiency of trains. The ‘car train’ would really require hybrid cars to work, and it turns out all you have to do is buy a hybrid car, and you’re already doing better than a train.

Maybe the ‘car train’ would still work, and would give you even higher efficiencies. But my original thought experiment was to see if there was something we could do with cars to make them as efficient as trains for inter-city travel. Turns out it’s easy. Just buy a hybrid, and you’re already there. No need for passenger trains at all.

I think the OP is on the right track (heh) but it could be more ad hoc. All cars on a highway should be able to convoy close enough to get the benefit of drafting even for small trips. Wireless communication and computer control can control braking and acceleration and control when cars join and leave the convoy. This would also descrease delays from congestion. We are not that far off from cars that can drive themselves on the highway.