The Cult of Cars, Racing and Everything That Moves You.
We may earn a commission from links on this page

A Physicist Explains The Three Biggest Challenges For The Hyperloop

We may earn a commission from links on this page.

We're fortunate in that we have a real physicist, Dr.Stephen Granade, on call to answer both my inane questions and sometimes more important stuff. This time we have something in the latter category. I sent him Musk's Hyperloop PDF and asked for his assessment. Here's what we got.

If you want to go fast, you've got to deal with all the things that want to slow you down. Two big ones are having to push yourself through the air and having to deal with the friction between you and the road or rails or whatever you're traveling on. The Hyperloop deals with both of these at once in a neat way. Unfortunately, I don't think it'll happen. I don't see us getting a Hyperloop any time soon.

First, let's talk about air. If you travel quickly, air piles up in front of you. The faster you go, the more the air piles up in front and the more resistance you get, which means you have to push even harder. And it's not what we physicists call a "linear effect". The faster you go, the worse it is. Bumping up your speed from 10 MPH to 20 MPH doesn't take nearly as much effort as bumping it up from 110 MPH to 120 MPH. It's why railway cars like the ones on the Shinkansen in Japan are so streamlined: to help the air flow over them and reduce how much piles up in front.

But if the air was moving as fast as you were, then you wouldn't have that problem! That's the first possible solution. Make a tube, fill it with air and your railway car, and then push both the air and the car through the tube at the same speed. That can work, but you've traded one problem for another. Instead of having the air pile up in front of the car, you've got this column of high-speed air moving along a tube, causing tremendous amounts of friction as it rubs against the walls. When I was a kid I'd sometimes get a Push Pop because my parents wanted my teeth to rot and fall out. If you've ever had a Push Pop, you know how hard it can be to shove the candy through the tube. That's the kind of effect we're talking about as the air moves really really fast in a tube.

Okay, then maybe we get rid of as much air as possible. We make a vacuum! A vactrain ( wouldn't have the friction and atmospheric drag of a regular train or a train in a tube full of air. Back in the 1970s, before maglev trains became a real thing, Robert Salter at the RAND corporation was pushing for these kinds of trains. But it's hard to get rid of that much air. Pumping a system down to 1% of atmospheric pressure is hard but doable. Pumping down to 0.1% is harder. Pumping to 0.01% is really hard. Pumping to 0.001% is crazy difficult. Just like pretty much everything else in life, it's the final percent that kills you. You have to get more and more expensive pumps, and because air is wanting to rush back into your tube, any leak or crack will kill your vacuum dead. If you want to get an idea of how hard getting a good vacuum is over any length of tube, take a look at what the Large Hadron Collider facility has to do to get just under 27 km of tube at a good, hard vacuum.… has more details about that than you could ever want.

Musk's approach with Hyperloop falls in between. Instead of pumping out all of the air, he wants to pump out most of it. He's talking about keeping the tube at about 0.1% of atmospheric pressure. It's about equal to the air pressure up at 170,000 feet. That's still a lot of air in the tube, though. It'll still pile up in front of the Hyperloop car, until you've got a column of air in front of the car that fills the tube and acts like the piston in a car. You're right back to having a lot of friction between the air and the tube, which will suck energy out of the system and keep the cars from going very fast.


Musk deals with that in a pretty dang clever way. He mounts a fan on the front of the Hyperloop car to suck that column of air into the car. That reduces air buildup and keeps it from hitting the point where there's a plug of air in front of the car slowing it down. Some of that air goes out the back of the Hyperloop car and through a rocket nozzle in a highly-engineered fart. Some of it is stored.

Some of it makes the car float.

The second problem you get with high-speed transport is friction between you and the road, where "road" can be an actual road or rails or cushiony magnetic field. Steel wheels on rails produce a lot of friction and heating. Maglev trains get around that by having the trains float on a magnetic field. There are magnets in the track and magnets in the train that repel each other.

Hyperloop would use "air bearings". It would float on a cushion of air like the way hovercraft do, or how the puck floats over an air hockey table. Some of that air would come from the air being scooped up in front of the car.

It's a neat idea, but would it work? The concept is sound, but I'm doubtful for a couple of reasons.

First, there's a huge difference between a neat idea that works on paper and a fully functioning system. Engineering is all about turning ideas into concrete things, and any engineer will tell you that what crazy people like me dream up often can't be manufacturered all that easily. The paper handwaves at the technical challenges of moving the cars from the low-pressure tubes into a station at regular atmospheric pressure. How good are your seals? How much air floods into the system every time you have a car come into the station? The proposed Hyperloop path goes through two tectonically-active mountain ranges. Will you have to blast huge tunnels through the mountain? Because a Hyperloop tube won't be able to curve sharply at all. If you're moving at 700 MPH, you don't want to turn very quickly at all, unless you like things like detached retinas.


And that's ignoring the real-world constraints of working with the federal government, state government, and every town and landowner between LA and San Francisco. You think it's hard to get permission to launch a new car? That's peanuts compared to the politicking that's required to get a long-distance mass transit system in place.

Second, while there are some neat ideas in Musk's white paper, the basic technology doesn't sound that different from the vactrains he dismissed so blithely. In his paper, Musk says that the problem with the trains proposed by the RAND corporation in the 1970s was that they required too low of a vacuum to be reasonable. But go look at Salter's 1978 RAND paper (…). In it, Salter says, "Cars travel in a reduced atmosphere of 0.1% of sea level pressure..." If that "0.1%" number sounds familiar, it's because it's exactly the atmospheric pressure that Musk is talking about in his Hyperloop paper.

The whole paper reminds me of the RAND paper, in fact. "Here's a neat idea. I've worked out a lot of the initial math. Now someone should go build it."

and that's the third big problem with Hyperloop. Musk isn't going to build it, not any time soon. He's got his hands full with Tesla and Space-X. He even told Business Insider as much (…) And Hyperloop would take someone with a lot of drive and resources to make it possible. The political barrier isn't insurmountable, but it would take a tremendous amount of time and energy to overcome, and that's leaving aside the technical challenges of turning this neat idea into reality. Without Musk's devoted attention, I see Hyperloop as being a lot like the 1970s vactrains: nifty ideas that have produced a lot of speculation and math but no real hardware.