Can The Hyperloop Actually Work?

Now I’m somewhat of an expert in all the technologies and I’m sorry, but I see some issues. First of all the booster vids.

The claims that Musk just came up with the idea are stretching things more than a little bit.  I first saw a vacuum transit system in Time Life’s Transportation book in their technology series back in the late 1960’s.  I suspect that it was referring to Robert Salter’s work, but I can’t be sure as I no longer have the book and my local library doesn’t either. It turns out, though, that the idea is even older than that, being shown in in Popular Science, in 1914.

Like a bad penny the concept keeps showing up.

Magnets Drive High Speed Suspension Trains Thru Air

The 400-Mph Passenger Train

I’ve always been a nuts and bolts type that likes to get into the details of machinery, especially transportation.  I collect stuff on how transportation works and just what needs to happen.  Plus, I’m always looking how systems work, whether or not it’s a commuter rail, high speed rail, shipping or trucking.  So I know how transportation systems are supposed to work and when things look wrong.  Unfortunately the Hyperloop, while it looks cool, feels wrong, very wrong.

I’m going to take this on in three parts. First is the details of magnetic levitation(maglev). Second will be the details of dealing with vacuum.  The third is operations  and what you would need to do to make Hyperloop pay and why in light of the other issues it just won’t work.

First of all, is hard. Very smart engineers in Japan and Germany have been working on maglev for as long as I have been alive, since the 1960’s and it’s still not competitive.  The problem is the huge amount of infrastructure involved in the Track way. Much of which involves things that are precise and need to be protected from the weather.  One thing that I have learned over the years is that precise, delicate and being outside in th weather doesn’t work very well.

The Germans have been testing and developing their Transrapid maglev for decades as the transportation system of the future. They haven’t been able to find any buyers in Germany due to the high costs involved and the only installation has been in Shanghai for a short line to the airport.

The biggest pushers of maglev have been the Japanese. The Japanese have been pursuing various forms of maglev technology since the 1960’s with varying degrees of success leading to the test site in Yamanashi.

Still the Japanese haven’t got the train fully operational.  If you look at the test schedule, there are frequent long gaps for maintenance and rework, which is normal for a prototype. Still the stuff on the RTRI and other sites doesn’t show a lot of operational details, so it’s difficult to get a handle on how operations run.  It’s hard to tell from the above video and the others I’ve found how often the maglev has to be recharged or what gas they are using to maintain superconductivity. If they are still using Helium, then things are going to be very expensive once real operations start. To say nothing of the leak and storage issues. If they have been able to apply high temperature superconductors, then liquid Nitrogen can be used and that makes operations much simpler.

It appears that the Japanese are continuing with liquid Helium superconductors.  I can understand that once they started down that road that they want to stick with it.  That does mean that they are dealing with liquid Helium and the assorted hardware, which has it’s issues.  I’ve worked in a place where we had lots of Liquid helium around and it’s A. slippery stuff that tend to leak even through solid walls and B. very expensive. That was here in the US where the only worldwide supply exists.  Also where there has been concern about shortages.  Depending on the number of trains JRCentral plans to run and how large the cryogenic system is that could be an issue.  The same goes for a Hyperloop and it’s magnets.  Which may not be able to escape the liquid Helium issues because of the next issue, the vacuum.

Vacuum sucks.  I’ve been working with vacuum in various ways for 18 years now off and on and I’m still discovering things.  It’s amazing how hard it is to keep the things you don’t want, out. The problem with dealing with vacuum is that you have to be always concerned with the stuff you put inside the chamber and any possible leaks.  It’s a given that there will be leaks.

While the Hyperloop doesn’t require a high vacuum it will be. by fat the largest vacuum chamber ever constructed if they build the track. Which means that it will need both large and numerous pumps.  Sort of like this one.

Here is NASA’s Space Power Facility which is apparently the largest vacuum chamber. which uses three of the pumps the size of the one above and still takes eight hours to pump down.  That’s because you can’t pump using compression for vacuum you have to remove the air, one slug at a time and as the pressure goes down, the slugs have less air in them.  So there is a long tail affect as it physically take longer for the air to get to the pump. In the Hyperloop’s case that means that you are going to need a lot of pumps running all the time in addition to powering the pods.

World’s Largest Vacuum Chamber

Now the Large Hardon Collider is probably the longest vacuum tube, but not the largest because it’s a very long pipe.  But the volume is small, because the pipe is probably 30 or 35 mm in diameter on the inside.  This raises the key point of vacuum design which is that you really want our chamber volume as small as possible.  With as few openings as you can get away with. Which is exactly the opposite of the Hyperloop’s requirements.

Then there is the issue of what happens during a catastrophic venting.  Which Thunderfoot’s video shows what can happen.

The Mythbusters did much the same thing with a ping pong ball.  For dramatic effect.

What will happen to the pod is shown by this video of a F4 being thrown at a wall. Which is a best case scenario.

Which brings me to the third and last set of issues, how operations would work or won’t work.  All The videos show one car at time moving down the tube. The problem is that if you did that, the number of people able to use the system is going to be very low.  Consider that the size of the pods that I have seen have a passenger load of 20 people or so.  So, if the pod takes half an hour, that means at best 40 people an hour, which is frankly ridiculous. Which means that you have multiple pods in the tube at the same time.

I’m not sure that is even possible because that requires, what is referred to in vacuum terms as a load lock.  Which would have to be pumped down in two minutes or so.  For a 3m dia x 20m load lock which is 141.4 m^3, from here,

You are going to need pumps that can handle about 2.5 m^3/second.  Which means that you probably need two of the bi pumps listed above.  Along with all the other pumps. All along the tube, all the tine.  And the liquid Helium.  Why, well all those magnets in the tube.  They will heat up and that heat has to go somewhere.  Which in vacuum means that the heat has to be bled out from inside the magnets because there is no air to draw heat away.

In any case the Hyperloop to work perfectly all the time or catastrophic and expensive failures are inevitable.  Add to that the every high costs  require just to keep the system running at all.  Would time savings be justified by most people?  Somehow I doubt that people are going to replace a $49 plane ticket for $200 ticket for being shot down the world’s largest gun.  This “cool idea” should go back into the “cool idea” folder and be put away like it was before.   It was unworkable in 1914, defeated by the laws of vacuum physics operational limitations and high costs and none of those have changed in the 100 or so years since.

Update: Here’s ThunderF00t’s video.







  1. MadRocketSci · July 22

    IIRC from my vacuum tech seminars: Vacuum pumps are rated by volume/time in terms of their pumping speed. Gas that is near vacuum is in “free molecular flow” regime. The atoms are all banging around and into the walls, instead of each other, so the collective behaviour of a fluid starts to disappear, and the gas stops responding to changes in density and pressure in remote locations. Instead, thermal speed means that a certain volume/time moves through any given cross section. A pump can handle some of that volume/time. Things like long thin pipes also produce “resistance” to gas flow (diffuse reflection from the walls), reducing the “conductivity” of the chamber->pipe->pump system. Elbows can kill flow rates too. You want any pumps directly pressed up against the chamber.

    So in our chambers, the base pressure we can pump down to is a function of the gas flow rate into our chambers, the volume of the chamber, and the pumping speed of our pumps.

    However, the hyperloop probably doesn’t care about quality of vacuum: All they would need is a very rough vacuum. But it will have to be maintained over time against leaks, which will probably require some power input proportional to the volume being pumped down. I’ll have to look up what they think they’ll do about the airlock-ing at the exit/entrance.


  2. MadRocketSci · July 22

    I dunno. I’m sceptical it will be profitable, not that it can be built, but I try to maintain an attitude of “let them try it.” To dig this country out of the hole it’s in, we need a lot of people building a lot of infrastructure. (We need nuke plants. We need oil shale development. We need our steel mills back. We need to rebuild all that reservoir pumping and water infrastructure if we don’t want California to turn back into the desert that the Spanish found!)

    Hopefully economically self-sustaining infrastructure for the most part. But without experiments, it’s hard to see how we can advance.


  3. David Lang · July 23

    > One thing that I have learned over the years is that precise, delicate and being outside in the weather doesn’t work very well.

    but with a hyperloop the maglev components are not out in the weather, they are inside the (partial vacuum) of the tube

    it’s hydrogen that leaks through solid walls, not helium. the low temps of liquid helium will make some materials brittle, but it won’t directly weaken them the way that exposure to hydrogen would.

    You don’t have to pump down a large volume to put a new car in the tube, the car just has to go through a tight-fitting airlock. If you take your 3x30m 141.5 M^2 car and give yourself a 2cm gap around the car, you now only have to pump out ~0.5m^2 of air (call it 1m^2 to allow for things like wheel wells and other recesses in the car)

    you are also missing how small an amount of air you are talking about being handled in one airlock compared to the volume of the tube.

    for a 400 mile run, you are taking ~650Km, or ~32000x the volume of your airlock (x16000 if you count the two airlocks and only have a single tube)

    your worries about items out-gassing in the vacuum is also misdirected. you are thinking in terms of lab or manufacturing vacuum where such things become contaminants to what you are trying to do, and that contamination corrupts what you are trying to do. In the case of the hyperloop, it’s not a succeed/fail thing. A bit too much air just reduces the efficiency/speed, there’s not a magic purity that is needed to make it work

    As for trading in a $49 airline ticket for a $200 hyperloop ticket. don’t forget that you are also changing the time drastically.

    Since you invoked Mythbusters, let me point you at the episode where they compare flying between SF and LA and driving between them, they were within 15 min of each other, so the real time is more like 6 hours vs an hour rather than 1 hour vs 30 min. With lots of small pods moving rather than aircraft, you don’t need to schedule your trip and get there early to make sure delays don’t make you miss your flight. You just catch the next pod.

    As far as catastrophic leaks, again I’ll point you at Mythbusters, specifically at the episode on explosive decompression and how they needed a pretty large explosion precisely placed, otherwise the result was not significant. In the case of a hyperloop, there is a huge volume to fill up, so the pressure will change slowly. As the pressure in the tube climbs, the cars in the tube will slow. you won’t get the ping-pong gun effect shooting cars down the tube for the simple fact that the cars are so much more massive. Yes, if someone were to blow a huge hole in the tube, the cars would be pushed, but the air would be leaking past at a pretty rapid rate (the more clearance there is in the tube, the less of an effect this would have, and since it’s not a high vacuum, they will want fairly good clearances anyway to avoid problems with compressing the air that will still be in the tube)

    Now, can it be done? who knows. There are a lot of things we can do now that we couldn’t do not very long ago. Musk has a pretty good track record between Tesla and SpaceX. I don’t know that I would bet my life savings on it, but I’m also very sure that I wouldn’t bet my life savings against it either.

    David Lang


    • jccarlton · July 24

      When the known unknowns in a project reach the size of a thick book, you really need to get your head out and start checking off the list. It’s taken the Japanese 50 years to check off the list on SC maglev and they are still not done.
      As for Helium, I was talking about how in cryo, the expansion ratio and the fact that He does bleed through just any possible leak means that gas loss is significant. Or all those trucks I kept seeing at the particle accelerator were not real. They went through about a truckload a week. A 600km system is going to require even more.
      There’s a big difference between explosive decompression where a low volume of air is trying to empty into a large(infinite) volume and implosion where the you have an infinite amount of trying to fill a relatively small volume very quickly. Pushing the pods along in front of it. There’s also the fact that if the pods are tight in the tube with a small margin of error, ANY pressure differential from a small leak could cause problems for the pod, in which case kinetic energy takes over and you have a rather large bomb, along with the cascade effect as the whole system shreds itself.
      As for pump down times and being able to operate at a worthwhile throughput, I chose what I considered a reasonable volume and the 120 second interval. Realistically the pump down rate needs to be even faster and there’s no pump that can physically pump down that fast.
      In any case, all this stuff and about 100 other things that I could list are going to make this more expensive, not less. The system will not run on solar power with left over power, the construction costs are going to ten times what they say with a huge amount of additional infrastructure not seen in the pretty pictures and any failure is going to kill a whole lot of people. It’s just not workable.


  4. Pingback: Looking At The Hyperloop, Part 2 | The Arts Mechanical
  5. Pingback: The Hyperloop Attracting Money, Getting Students To Do The Engineering | The Arts Mechanical

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s