Is Elon Musk’s Hyperloop Really an Alternative to HSR?

Jul 15th, 2013 | Posted by

Elon Musk is getting attention again for his proposal to use tubes to connect San Francisco to Los Angeles in 30 minutes. I was more than a little skeptical when I first wrote about Musk’s concept last September. Today we’re learning a little more about the “Hyperloop” and how it might work in practice:

For a while now, Musk has been hinting at an idea he calls the Hyperloop—a ground-based transportation technology that would get people from Los Angeles to San Francisco in under half an hour, for less than 1/10 the cost of building the high-speed rail line. Oh, and this 800-mph system would be self-powered, immune to weather, and would never crash.

What is the Hyperloop? So far Musk hasn’t gotten very specific, though he once called it “a cross between a Concorde and a railgun and an air hockey table.” But we’ll soon find out more. On Monday, Musk tweeted that he will publish an “alpha design” for the Hyperloop by Aug. 12. Responding to questions on Twitter, he indicated that the plans would be open-source, and that he would consider a partnership with someone who shared his vision. Perhaps the best clue came when he responded to an engineer named John Gardi, who published a diagram of his best guess as to how the Hyperloop might work:

And here’s Gardi’s conceptual graphic:

So it’s not exactly the same thing as the tube transportation system from Futurama, but it’s close. I’m all for very fast methods of transit, and downtown SF to downtown LA in 30 minutes would be a game changer. But there remain some practical problems facing the Hyperloop.

Musk has claimed that it could be built for just $6 billion. If the above design is anything close to accurate, and Musk says it is, then I have a really hard time seeing how it can be built at that cost.

In order to reach speeds of 800mph, those tubes will have to be in a straight line. Will it have to be completely straight all the way from SF to LA? I am not sure, but neither will the system be able to have many curves, either horizontally or vertically. In California, that means tunneling. The cost of tunneling alone will far exceed $6 billion. Perhaps Musk has some sort of solution to this, but if you are going to gain elevation or have any curves in the tube then there’s no way to maintain the 800mph speeds you need to cover the distance from SF to LA in 30 minutes. And as we’re seeing in Beverly Hills, tunneling isn’t always popular with the locals.

Further, the cost of materials and labor will also help drive the cost up above $6 billion. Building a series of tubes is not substantially different from laying tracks. And if much of it has to be in a tunnel, the cost will skyrocket rapidly. Depending on the location of the tubes, you’ll see the same NIMBYs who have spent five years trying to kill the high speed rail project work hard to kill the Hyperloop.

Some HSR supporters are speaking well of the Hyperloop:

But it would be a mistake to dismiss Musk’s seeming far-fetched notion, said Rod Diridon, executive director of Mineta Transportation Institute at San Jose State University. “Mr. Musk has a reputation for state-of-the-art problem-solving, most clearly shown with the Tesla,” Diridon said. “We look forward to helping mature his concept to a practical application as quickly as possible.”

That might not be fast enough for California. Diridon detailed the lengthy process a transit solution this extensive would have to navigate, from land-use approval, funding, development to construction. “The problem is that this can’t be built within the time frame the state needs in order to meet the huge influx in population over the next 25 years,” he said.

Diridon spoke highly of Musk, and said he plans to buy a Tesla Model S when the lease on his electric Nissan Leaf is up.

“It’s a good thing, and we need to have it as the next move beyond high speed rail. But if we try to use it instead, we’re going to have nothing in the middle of a population influx like no state has seen in the past,” he said. “I’m not critical — I’m practical.”

Diridon makes sense here. I’m all for figuring out how to make a Hyperloop work. But Diridon is right when he points to these obstacles, in addition to those that I’ve cited. HSR is a solution that is ready to go right now, that will break ground this year. A Hyperloop isn’t a competitor with HSR, it’s the next evolutionary step that is also not nearly ready for prime time.

I don’t quite know why, but a lot of people find it hard to accept that it’s high speed rail which is on the cutting edge of transportation technology. It’s energy efficient, usually makes an operating profit, and is a good use of space with minimal impact on the surrounding environment. Steel wheel high speed trains have shown innovation and technological advances, with France testing a ultra high speed train in 2007 at 356 mph.

Any number of investments we’re making today could be rendered obsolete by some massive leap forward in technology. But it’s hard to see where that comes from in transportation. High speed rail meets all the challenges we’re likely to see over the next decades. It’s a proven success around the world. And it’s proven adaptable and expandable, able to keep up with new technologies.

California should continue to invest in innovation and bold new ideas. It also needs to invest in sensible, functional, proven solutions that work well around the world. Maybe tubes will be the basis of transportation someday. But for as far as I can see, high speed rail is where California’s transportation future lies.

  1. Useless
    Jul 15th, 2013 at 20:50

    It all comes down to cost. Anything other than steel rail HSR is simply not affordable, including the maglev Chuo Shinkansen.

    VBobier Reply:


  2. Donk
    Jul 15th, 2013 at 21:03

    Self powered? And what about air resistance and friction? Or is this a vacuum? A self powered vacuum?

    Clem Reply:

    It has to be a vacuum. The friction of a 600 mph air column on the walls inside of a pipe like this would be off the charts.

  3. D. P. Lubic
    Jul 15th, 2013 at 21:07

    My first thought is Robert’s–this thing would essentially be all tunnel. How can you build all tunnel for $6 billion? How are you going to deal with the earthquake faults on the route?

    A second question involves heat. This design moves a lot of air around the loops. This is in a tunnel that’s hundreds of miles long. For the record, the Channel Tunnel has to deal with heat build-up from motors and from pushing air through the tunnel with a cooling system. It’s a conventional railroad not moving all the air (and dealing with the air friction on the tunnel walls) that this system will have.

    Finally, there are no intermediate stops. I know some complain that the proposed California HSR system has too many stops (and a big “detour” as well), but do you want to build a system with no intermediate stops at all?

    synonymouse Reply:

    @ D.P.

    When I hear this scheme I start thinking Howard Hughes.

    But apparently Hughes “transcendental” period began when he hit his head really, really hard on a test flight.

    For gadgetbahn hustlers I guess it is back to monorail and skybus.

    D. P. Lubic Reply:

    Oh, as to how it works, it’s essentially a pneumatic tube for people, like what used to be used in department stores, and like the very first New York City subway. Main difference is using magnetic propulsion and braking for acceleration and deceleration into the air stream or air column.

    D. P. Lubic Reply:

    Ought to include this page for reference; couldn’t find anything like a pneumatic system while looking, but I was short on time, there might be something there. It certainly has a lot on maglev and PRT:

    D. P. Lubic Reply:

    D. P. Lubic Reply:

    Not quite as fast, but pretty good as it is; note that this locomotive was built to test passenger cars:

    Andy M Reply:

    also Google something called Swissmetro.

    Max Wyss Reply:

    Yeah, the Swissmetro; it comes up every few years and gets dumped as quick as possible. Its cost is (if I remember correctly higher than those 6bn USD, just for the approximate 120 km between Zürich and Bern.

    To stay with pneumatic trains, the Great Western Railway had a pneumatic system between Exeter and Newton Abbot (along the famous Seawall) for a few years; it was kind of successful. “Power” cars had a piston in a tube, and there were (to be expected) seaing issues.

    Alon Levy Reply:

    How can you build all tunnel for $6 billion? How are you going to deal with the earthquake faults on the route?

    Using the magic of startups. I mean, if you could start Microsoft in a garage, why can’t you build all-tunnel HSR for about $10 million/km?

    Andy M Reply:

    A tunnel need not be an underground tunnel. In areas where there aren’t many NIMBYs you could make it a pipeline. You wouldn’t know from inside and it wouldn’t change the operating principle.

    Alon Levy Reply:

    It requires tunnel lining.

  4. John Burrows
    Jul 15th, 2013 at 21:26

    Sounds like Hyperloop is an exciting possibility for a future transportation system—A system that could possibly be carrying passengers by the time the first space elevator become operational. In the meantime we have a CAHSR groundbreaking on the schedule before the end of summer.

  5. EJ
    Jul 15th, 2013 at 22:32

    The advantage of high speed rail is that it isn’t “on the cutting edge of transportation technology.” It’s been around for over 50 years. So building out a system in CA should be a no-brainer.

    trentbridge Reply:


    Ben Schiendelman Reply:

    +1 !

    VBobier Reply:

    This musk vac-tunnel idea is just that, an idea, it’s also untried and unproven anywhere in the world, while HSR is tried and proven successful worldwide.

    Joey Reply:

    The idea of VacTrains isn’t anything new either. The concept has been floated in the past and deemed impractical with today’s technology and industrial output.

    VBobier Reply:

    You won’t get disagreement on that from Me.

    EJ Reply:

    I’d put it in the same category as fusion power – a basically sound concept, with a bunch of pretty significant practical hurdles. Definitely worth investing in research, but not anything I’d want to hang my hat on for a practical solution in the next 50 years.

  6. Clem
    Jul 15th, 2013 at 22:52

    I’m curious to see what he’s got up his sleeve. How do you switch routes? How do you ensure survivability in case of vacuum breach? How do you keep trains from smashing into each other in case of unplanned emergency stop, since you clearly can’t maintain the traditional separation of one braking distance + margin? I think this will all revolve around the safety aspects, just like it does for airliners or any other safety-critical, highly engineered product.

    D. P. Lubic Reply:

    I think I can answer part of this:

    Survivability of Vacuum Breach: Non-existent problem, the tunnel uses positive air pressure to move cars or trains.

    Maintaining Stopping Distance: My guess is that the system would be relying on that positive pressure to keep trains or cars separated. Think of how the air pressure behind a car not only moves the car, but the air in front of it, and thus also is pushing a car ahead as well.

    Main concern is whether power consumption might be high. That’s a lot of air to keep moving around.

    Switching will be interesting, too. I would guess the suspension will be some sort of maglev for running at the anticipated speeds. This would be suspension only, not traction, power to move coming from positive air pressure provided by blowing engines (yeah, I know, old terminology).

    D. P. Lubic Reply:

    This of course is assuming the descriptions are right, and we aren’t actually working with a vacuum system, as Clem suggests by the potential difficulty of moving air at 600 mph through a pipe.

    thatbruce Reply:

    @D. P. Lubic:

    In the event of a tunnel breach occurring between two cars, the lead car no longer has positive pressure behind it (because it’s all going out through the breach) and slows down. The following car, still with the positive pressure behind it, zooms past the breach and into the slower-moving lead car. You’d need an active braking capability on each car to avoid this issue.

    He’s asserted that it’s not a vacuum tube though.

    VBobier Reply:

    Don’t forget that air can be compressed, so I don’t see air as something that can effectively keep trains separated, the more air one has to move, the more power one will require, on paper everything looks good, in practice ideas that looked good on paper don’t always pan out as anything other than fools gold…

    Alan Kandel Reply:

    This one uses the pressure differential principle to move a piston inside the cylinder (power tube) located in the truss situated beneath the train. Through magnetic coupling, the train running on rails on top of truss is magnetically coupled to the train.

    Alan Kandel Reply:

    Make that, the movable piston is magnetically coupled to the train.

    jonathan Reply:

    .. So many of the posts here view a “pressure difference” in a non-scientific, even pre-Archimedean way.

    if the vehicles are being propelled by a “pressure differential”, that “pressure differential” has to go _all the way_ back to the starting point. More, the pressure differential has to exist all the way to the end of the tube. Think of it as a big blow-pipe pushing a dart; the dart is the capsule containing passengers.

    Send one dart, just one, down this ~800km tube, just _one_, and you now have a big fat column of air all the way down the tube. (If you don’t have big fat column of air all the way down the tube, then the pressure-differential is doing bugger-all work to propel the capsule.) That ~800 km column of air means the next capsule is not in a vacuum, and therefore is NOT’t going to be travelling down the tube at 600 mi/hr. Mach 0.8 with all that luverly ground-effect from the tunnel walls?!?

    Is basic science education in the USA _really_ this bad?

    thatbruce Reply:


    Pressure reservoirs beside the tunnel at frequent intervals that suck air out of the path of an approaching capsule, and push it back in very quickly immediately after the capsule passes. Think of a sealed giant piston protruding from the outside of the tunnel that retracts to create a partial vacuum in the tunnel, and expends afterwards. This isn’t a serious answer as you either need a lot of these stations, or they need to be large in order to move a lot of air very quickly, both of which would have a lot of impacts on the cost of the system.

    JB in pa Reply:

    I was thinking of a ducted fan in the car injecting air between the car and the wall with the side of the car shaped in such a way as to eject the pressurized air out the back at near the travel velocity to control the slipstream between the moving volume of air and the wall. There are options to consider as to where the fan inlet would be located; at the front or the back.

    JB in pa Reply:

    The above comment refers to a system pressure of between 2 to 4 psi. About the same as an airliner experiences at FL400. (Repost comment in correct location due to iPhone )

    JB in pa Reply:

    Thinking of slipstream control I am not sure if the air should be injected between the car and wall or extracted. Or if the jet velocity should be the travel velocity or half the travel velocity. The general idea being a configuration that avoids unwanted boundary layer effects. The system should be self correcting and stable. Reminds me if the micro aerodynamics of a hard drive read write head. Or the aerodynamics of a race car.

    JB in pa Reply:

    Extracting air from the side wall is interesting. The shape of the leading edge, wall, and trailing edge of the car may be important. Each car would move along the wall efficiently and leave in its wake conditions beneficial for the following car. It may be a requirement that a number of cars travel together or that the spacing of cars maintain a max min continuous pattern.

    JB in pa Reply:

    Maybe each car could gather any turbulent air and initiate laminar flow. The car would be short enough to be within the laminar flow before the air between the car and wall becomes turbulent. The slipstream is ejected behind the car to minimize the disturbance between the moving slug of air and the static wall between cars.

    JB in pa Reply:

    The air injection or extraction system is in the car not a distributed complex system in the wall.

    JB in pa Reply:

    Make such a tube system between some of the Hawaiian islands.
    Use for passenger or maybe better a dedicated freight system.

    JB in pa Reply:

    Maybe Elon has chosen too high a velocity whether it is passenger or more if a freight system.

    JB in pa Reply:

    Re: above – To clarify I meant – slower for a freight system.

    Robert Reply:

    Take a look here, at the FAQ:

    thatbruce Reply:


    Maybe he’s talking about using air pressure to suspend a Concorde-like capsule inside a large diameter rail gun, accelerating the capsule up to speed and sending it on a sub-orbital trajectory. Maybe people will be ensconced in large pucks and Norse gods will be summoned for a trans-California game of air hockey.

    JB in pa Reply:

    The above comment refers to a system pressure of between 2 to 4 psi. About the same as an airliner experiences at FL400.

    JBaloun Reply:

    Aside from 14.7 psi or 0 psi there are other options, like 3 psi. +\-. This may reduce the heating problem.

  7. Michael
    Jul 15th, 2013 at 23:25


  8. Ben Bethel
    Jul 15th, 2013 at 23:52

    Perhaps Musk sees CAHSR as a competitor to the Tesla line of vehicles and just wants to create some confusion at the moment? Perhaps he sees autonomous vehicles as having a potential for 8 people to share the expenses of just one vehicle, reducing the overall demand for cars by quite a bit. We have to prepare for a California with 80 million people in 40 years and 100 million in 65 years… like it or not, it’s going to happen. It’s really crazy to think, but we could sustain 80 million people in the state without even building another house since many of us live in rather spacious homes with room for 2x the people they have today. Fewer of us need to drive to a job. More of us will live closer to our jobs. We really will be sharing cars, so it could be possible that a state with 2x the population one day could have 1/4 of the cars on the road… just ranting, it’s late… but this is a post that is just begging for people to rant. I’m very interested in seeing this plan… but it’s being hyped up like the Segway was… and if the Segway wasn’t over-hyped and had a cooler, hipper design and was a fraction of the cost, then they’d be really popular today, because they really do offer a solution to short-distance commuting…

    Ben Schiendelman Reply:

    I think you’ve hit the nail on the head. It’s like the highway industry funded BRT people.

    synonymouse Reply:

    The highway lobby has always been the natural enemy of rail – it is a jihad with them.

    Very difficult to predict where tech is headed or will break thru, but quantum computing and AI look pretty robust and promising. Supposedly existing high-end computers achieve about the i.q. level of a 4 year old human.

    Vast improvements in computing could make presently bogged-down undertakings possible. In particular nuclear fusion. Ditto for room temperature superconductivity and maglev type schemes.

    And i’d suggest aviation innovations are more likely than vactrains.

    VBobier Reply:

    Vast improvements in computing will have to get beyond the current limits of silicon, which is nearing it’s limit… So far there have been proposals, but not much else.

  9. swing hanger
    Jul 16th, 2013 at 00:54

    Swinging for the fences when diligent, small ball will win you the game with more certainty.

  10. Emmanuel
    Jul 16th, 2013 at 01:03

    That would mean he would only have to spend $7.5 million per mile. That wouldn’t even cover property. Hahaha, it’s a good laugh though.

  11. ericmarseille
    Jul 16th, 2013 at 04:16

    If I were an American-hater I’d say “go for it, guys! Jump on that unique chance for the US to soar from High-Speed-Rail laggard to uncontested leader!”.

    and then I’d chuckle at the trillion or so that thing would cost to build and run…

    BTW I’m having real doubts on Mr Musk’s other enterprises right now, esp. his Supercharger network (which is in itself an excellent idea, don’t get me wrong)…Free, for life, really? That man must hope not too many people buy his cars and not too many superchargers get installed too early…He’s really walking a tight rope financially in my opinion.

    synonymouse Reply:

    I have been watching the France 2 jt coverage of the wreck at Bretigny-sur-Orge and am fascinated by all the hand work on rebuilding the switches.

    I know next to nothing about rail maintenance but it just seems that air socket wrenches with a fixed torque setting would be more reliable and more predictable. It would appear that the guys are just feeling the right level of resistance on those long hand wrenches.

    One thing is for sure – something caused those bolts to come undone and that plate to travel and lodge in the pointwork.

    thatbruce Reply:


    Switches are hard to do automated maintenance on, and if they’re planning to bring the damaged switches back into short term operation and replace them with a pre-built switch in a month, then hand tools and slow speed orders are the way to do it.

    ericmarseille Reply:

    I’ve looked for technical info regarding that horrible crash :

    – the train itself (which didn’t fault) had been controlled by the middle of June
    – the fatal switch had been controlled lastly on the 28th of June!!!

    Which means that either the switch was controlled by one of these nonchalant, goofy, lazy pricks that generation Y has been producing by the millions (you know, those “Doctors es Playstation” who can hardly speak their own language), or, ever scarier, that it was put out of place on purpose, which, given the number of hostile muslims, including in the SNCF employees, or simply the number of loony fucks we have now in this once livable country, is highly possible.

    One of the worst rail catastrophes of the 25 last years or so was caused by a lunatic who simply put a big metal piece on the railway to “see what it could do to a train running over it”…

    Richard Mlynarik Reply:


    Thanks for doing your part to make the US look good. George Zimmerman salutes you.

    I’m sure Al-Qaeda is lurking somewhere here.

    swing hanger Reply:

    Eric, here is the trackwork in question. Does look like either pure neglect or purposeful sabotage:

    EJ Reply:

    Interesting that despite being full of right-wing knuckle draggers like this one, France did manage to build out a world class high speed rail network.

    ericmarseille Reply:

    Ouch…My heart bleeds (sincerely).

    I’ve been accused of being racist, leftist, socialist, but never ever a “right wing knucke dragger”.

    And I just hate what Zimmerman did to that boy (caught it through Bill Maher on YouTube), and I hate guns galore, and I hate violence, and I hate hypocrisy, and Burqas, and “reasonable arrangements”, and phallocracy.

    And before you tell me “seems you hate logic too”, please live the life I’ve lived.

    synonymouse Reply:

    I suggest there are a lot of Americans who imagine Europe as a sort of paradise from our problems but in reality “mean streets” are everywhere now.

    My impression from the news reports was that the first responders to the Bretigny-sur-Orge wreck and fatalities were pelted with rocks thrown by gangs of locals.

    adirondacker12800 Reply:

    Ya have a cite for that? and not one from some conspiracy theorist’s web site.

    ericmarseille Reply:

    It’s alas true. So revolting. The “locals” were an informal bunch of thugs who came to rob the wounded and the dead (!).
    When the stunned rescue teams understood that they didn’t come to help and told them angrily to leave the scene, they responded by stoning them.

    This country is simply going bonkers and we French people of all origins (the thugs don’t even consider themselves French, they call us the “Céfrans”, “verlan” slang for “French”) are fed up with that violence and lack of respect for all kinds of institutions, for all kind of people, and all ages.

    synonymouse Reply:


    From watching the France 2 news stories it would appear that the non-hsr lines have fallen behind in maintenance. They showed some images of trackage in pretty poor shape.

    I have to assume this is a budgeting issue. I have brought up this problem in relation to the CAHSR, that is if they insist on an expensive to run and maintain, circuitous route that loses money the old curse of deferred maintenance will surely fall upon us. The DogLeg is a mistake. You have to go with the most direct alignment. It is the only one that can generate the revenues necessary to guarantee adequate maintenance.

    synonymouse Reply:

    @ ericm

    A better American equivalent for “controlle” is “inspected”. Sorry about the accent marks or I mean their absence. I am afraid to mess with my keyboard settings, which tend to go crazy on their own if I punch a wrong key.

    ericmarseille Reply:

    Well thanks synonymouse! I should’ve remembered! Feel always welcome to correct me…

    Alon Levy Reply:

    That’s what happens when you tell a web entrepreneur that he’s a general-purpose genius who can reinvent everything. He’ll actually believe it, and the media will keep talking about his ideas because hey, he made money doing something completely unrelated, so he obviously has a clue. Meanwhile, stories about the actual, proven low construction costs of Spain are nowhere to be found in mainstream US media – too boring, and Everybody Knows Germany is Better.

    Neville Snark Reply:

    Have you heard of Dysonrail?

    Keith Saggers Reply:

    Yes, and I have a bridge to sell you

    Paul Dyson Reply:

    Actually DysonRail is the name of my logistics company. But thanks for thinking of the family sucking business. I’m surprised no one has mentioned I.K. Brunel, George Medhurst and others from 1799 onwards who have tried to make sense of this idea. You can see a track section of Brunel’s South Devon Atmospheric Railway at Didcot in the UK.

    jonathan Reply:

    .. it’s such a bummer when the rats eat through the leather of your leather pneumatic hoses…..

  12. TomA
    Jul 16th, 2013 at 07:46

    Even IF they made this, you would assume that it would ONLY go from LA to SF.

    You would still need something in between to connect the smaller makets.

    The real thing that would be if this could be made to work (and of course it won’t) it would be perfect for the backbone of a transcontinental network.

    So you would have Hyperloop connecting major cities. HSR connecting mid-szied cities to the major cities. And traditional amtrak style rail connecting those cities to even smaller cities with buses for the final leg to small towns.

  13. Reedman
    Jul 16th, 2013 at 09:19

    I guess the economics are based on having no land acquisition costs except for the terminals (the same reason air travel works well, and why passenger rail systems can only be done by the government). Like high voltage DC power distribution, the proposal probably works well for connecting endpoints (LA to San Jose), with no intermediate taps/stops.

    It never hurts to get some new thinking on a problem. I look forward to Musk’s presentation.

    Peter Reply:

    It’s not the land acquisition costs that drive the costs of major projects, it’s the actual construction costs. Everything else is pretty much pocket change.

  14. JJJ
    Jul 16th, 2013 at 11:44

    Theres an easy solution to concerns about curves, tunnels, elevation, property etc…

    …..the ocean. Its perfectly flat, has no NIMBYs, and with some floaties, your only costs are the tube itself. And emergency life rafts. .

    Matthew B. Reply:

    The California coast is curved. Draw a straight line between SF and LA and it goes over land. Either the tube would curve, or it would be a few hundred miles out to sea to get to it.

    Also, it seems that no cities in between would be served. Sorry, Silicon Valley, Central Valley, and San Fernando Valley, Elon doesn’t want to stop for you.

    synonymouse Reply:

    Still Palmdale-possessed are we?

    Peter Reply:

    What? Pull up Google Earth (or hell, an atlas) and draw a line between SF and LA. It does not touch the ocean, unless you count the SF Bay as “ocean”. Interestingly, though, a straight line between SF and LA does pass straight through San Jose.

    adirondacker12800 Reply:

    Maps are distorted to display curved surfaces on a flat screen.

    Great circle route between SFO and LAX

    Clem Reply:

    Wow, straight through Tejon Pass!

    JJJ Reply:

    A tubed system can handle curves, just not too many sharp ones. Look at your bank pneumatic delivery system at the drive-thru. An ocean line could have very wide, gentle curves, and never change in elevation, vs anything on land.

    thatbruce Reply:


    the ocean. Its perfectly flat, has no NIMBYs

    Just some pesky Marine Life Protection Act supporters.

    JJJ Reply:

    Tubes dont disturb marine life.

    Joey Reply:

    Installing them does.

    thatbruce Reply:


    When the MLPA debate was in full swing a few years ago, the rhetoric used on both sides would have you believing that merely dipping your toe in the water would cause so much pollution that commercial fisheries would have no choice but to enforce a strict cordon.

    VBobier Reply:

    Some forget about the CA Coastal Commission, it would have final say about anything that impacts the coast.

    Eric Reply:

    The ocean floor has plenty of curve and elevation issues. The ocean surface has ROW issues (boats).

    I suppose you could suspend a tube a certain distance below the surface (lower that the deepest tanker/aircraft carrier). But the cost would be very high, as would the risk of catastrophic flooding.

  15. MarkB
    Jul 16th, 2013 at 16:01

    The ET3 site talks about how large a chuck of the world’s transportation needs it can solve. According to the site, standard throughput per airlock is about 800 people per hour. However, an HSR consist can seat 1,000 people and depart with 5-minute headways. How does ET3 equalize capacity? Does it build a dozen-tube “tubeway” to handle the great-circle trunk line? And what would that do to land acquisition, construction and maintenance costs? I just don’t see how it achieves the passenger throughput it would need to be viable, let alone transformative.

    The site talks about using the freeway interchange model, but that brings a whole other can of safety and operation worms when you start talking about adding and subtracting live capsules moving anywhere from 350-4,000 mph with 26-second headways (or less). Sounds like an awfully low margin for error in timing, switching and acceleration/deceleration. The site dances around this by saying “automation reliably transcends human limitations.” Uh, right. Computers have no bugs, no siree Bob.

    On another matter, capsule occupants are in a semi-prone position with no room to stand. Even for a 45-minute trip to NY, what do you do about bathroom or food needs, let alone latent claustrophobia?

    I hope Musk has something else in mind.

    Robert Reply:

    “How does ET3 equalize capacity?”

    If I read the site correctly, they do this via multiple airlocks feeding into the main tube. Each airlock does 6 people in a cab every 26 seconds. If the tube design is for 350mph, they claim the tube can handle 10 capsules per second. That works out to 200,000 passengers per hour.

    After having read the entire FAQ, a couple things strike me. There is no doubt that modern computers can handle 10 capsules per second traveling around the system. The mechanical aspect is what I would worry about. Many things can go wrong, and they cover a lot of these in the FAQ. But the “bad guys” just need to slice the tube in a single location, and all of a sudden you possibly have 60 people per second shooting out of a rail gun into the air at 350mph.

    Of course, you could also build an intrusion detection system consisting of sensors to detect if anyone comes close to the tube. But you still have the risk of someone flying a plane into the tube(s).
    I would think a first application for something like this would be where you have a high capacity point to point need for moving sat moderate weight cargo. The ports of LA and Long Beach come to mind. There was an idea floated a few years back to put a point to point maglev link between the port and San Bernardino. That would avoid congesting the LA basin with the current trains and/or trucks and the associated pollution and traffic. The ET3 would be that on steroids. And if they could build it as cheaply as they state, it would be way less expensive than maglev, and might work very well in that application.


    MarkB Reply:

    I read that “10 capsules/second” bit but don’t think it would ever work operationally, because that leaves basically zero margin for error in a collection of mechanical systems moving several-hundred pound slugs into and out of the way. All it would take is for one capsule to accelerate a weeee bit faster than the others (say, because of a weight difference) or for a mechanical gate release to take a fraction of a second longer than expected to do its thing) and your 10 capsules/second start becoming too friendly with each other. Or, consider a long, gradual uphill to cross the Continental Divide: wouldn’t heavier capsules slow more than lighter ones only 1/10th of a second away? Remember, they’re just coasting, so gravity and inertial will have their way. Or think of the mechanism at an “interchange” that has less than 1/10th of a second to switch tube paths an earlier mainline capsule and the target capsule to be shunted, then less an another 1/10th second to switch the tube path back for the next mainline capsule. That’s an enormously complex and sophisticated mechanical system! And we’re not talking about reliability factors suitable for boxes of Chinese junk, we’re talking reliability factors for human transport! No, I see that 10-capsules/second claim as operationally bunk.

    Eric Reply:

    “wouldn’t heavier capsules slow more than lighter ones only 1/10th of a second away”


    More precisely, the heavier capsules would slow slightly less, being less affected by air resistance.

  16. Reality Check
    Jul 17th, 2013 at 21:58

    Will Elon Musk’s Top Secret ‘Hyperloop’ Put California’s High Speed Rail Out of Business?

    California transportation officials, who have yet to break ground for the long-awaited, ultra-expensive high speed rail line, must be getting nervous.

    Musk has said that he was inspired by California’s high speed rail project — but wanted to do it better and at less cost. The billionaire described that project as the “slowest bullet train in the world, and the most expensive per mile.”

    Joey Reply:

    Musk really has no clue what he’s talking about. Ignoring the flaws with his proposal (how exactly do you plan to build untested VacTrains for $10m/mile?), he gets some basic facts wrong – (1) CAHSR is rather expensive by world HSR standards but not the most expensive – HS2 is managing to cost more, for instance and (2) CAHSR will manage respectable average speed – upwards of 250 km/h even in the worst scenario.

  17. Joe
    Jul 25th, 2013 at 23:14

    I love this blog, read it regularly. Big supporter of HSR, and I agree on going forward with existing technology until a new tech is proven. That said, Musk is a genius. He’s not an idiot who would overlook obvious problems like tunnels, and he’s not going to unveil this thing unless he has a legit plan. We have to wait a month, but I’m sure his alpha plan will address ROW and tunneling. As for intermediate stops, this design lets you have as many as you want and each individual tube will be non-stop. Also, I hear people accusing Musk of undermining the rail project because he’s a car maker — while the big automakers would stoop to that level, interviews with Musk make it clear this guy is about doing big, awesome things, like space colonization, autonomous electric cars and solar energy. He’s on the side of progress, like us, even if he puts down HSR.

Comments are closed.