The overhyped hyperloop: is Elon Musk's plan a superfast load of nonsense?
The Hyperloop as pitched – as a $6bn project usefully linking two cities – can't exist. But the excitement around it could destroy any hopes for a workable Californian high speed rail link.
It's hard, in this day and age, to properly earn the title "eccentric billionaire", but Elon Musk, the 42-year-old South African who made his fortune as one of the founders of PayPal, is getting there. Since the company to eBay in 2002, Musk has been prowling for other ideas which will change the world. Given the most conventional of those is completely retooling America's private transportation infrastructure with the most serious push yet to popularise the electric car, he's certainly earned the "eccentric" tag.
As well as Tesla Motors, which has so far brought two cars to market and installed a network of high-speed chargers up and down the east coast of the United States, Musk also runs SpaceX, a private space transport company. Like Tesla, this is no pie-in-the-sky concept; last year, SpaceX became the world's first privately held company to send a cargo payload to the International Space Station, carried into orbit on the back of one of its Falcon 9 booster rockets.
Now, Musk has a third idea. He's too busy to actually do it – running three companies at once is apparently too much even for him – but he's bequeathed the idea to the world, in the form of a 57 page PDF: the Hyperloop.
Described by Musk in the hype campaign building up to the launch as a "cross between a Concorde, a railgun and an air hockey table", the Hyperloop is intended to be a revolutionary new form of public transport. Capsules containing 28 passengers each would be accelerated up to 760mph through what is, in essence, a giant railgun. They then travel the distance between LA and San Francisco in a low-pressure tube, floating on a cushion of air, before being slowed back down at the other end just half an hour later.
It's futuristic as hell, but if you read the "alpha" proposal, Musk thinks it can work. He writes that "short of figuring out real teleportation, which would of course be awesome (someone please do this), the only option for super fast travel is to build a tube over or under the ground that contains a special environment."
For something which is barely more than napkin scribbles – the proposals describe it as "an open source transportation concept" – there are an impressive number of details worked out already. In order to save on land use, the Hyperloop would run on pylons above the main motorway between the two cities. The air on which it floats is sucked in from the front, in order to avoid the "Kantrowitz Limit" (think about the problems in trying to shove a plunger down a tube really fast). The battery packs running the fans, and the linear electric motors which power the cars, are charged largely by solar power, plentiful in sunny California. In order to reduce energy use even further, regenerative braking is used at each end, converting some of the kinetic energy back into electrical power. There's even an estimate of how well the tube would cope with an earthquake or terrorist attack.
But for every detail worked out thoroughly, there's one which is slightly glossed over. Take the route: throughout the plan, it's described as being from "San Francisco to Los Angeles". But that's not strictly true.
The Los Angeles stop is just south of Santa Clarita, a 40 minute drive from downtown LA. That's bad enough for a city where the main airport is very much in the centre of town, but it's worse when you consider that Musk envisages the system being used "more for commuting than for vacations". Public transport in LA isn't great (the same trip would take around 2 hours on a mixture of buses and metro rail). Perhaps Musk envisages his commuters taking two taxis every day, in addition to a return trip through a vacuum tube?
The situation in San Francisco is more ambiguous. One map shows the tube crossing the Bay Bridge, but a second shows it terminating in San Lorenzo, again a half hour drive from San Francisco proper. At least San Francisco has working trains, though; a BART would take you into town in the same time (Again, however, the airport is closer in to town).
So while it's not untrue to say that the system makes the journey from San Francisco to Los Angeles in 35 minutes, the useful journey time is considerably longer. That one guy who lives in Santa Clarita and wants to take a job in San Lorenzo will be over the moon, though.
There are similar elisions elsewhere. The cost of land is the second biggest single outlay, estimated at a billion of the six billion dollars Musk budgets; but that assumes that land already owned by the state – specifically, the median of the I-5 interstate – is free. In practice, the California transport department might have something to say about that. And even if they don't, "using existing resources" isn't the same as "using no resources". Building a hyperloop on the central reservation of a motorway necessarily prevents other things being built there, such as an extra lane for traffic, the midpoint of a footbridge, or even just a more conventional mass transportation system.
Whether or not the hyperloop even is a mass transportation system is probably the biggest thing of all. The draft proposal says that:
Assuming an average departure time of 2 minutes between capsules, a minimum of 28 passengers per capsule are required to meet 840 passengers per hour.
In rush hour, the departures would be sped up, allowing slightly over 1,000 passengers to be carried in an hour, and the theoretical maximum of the system is around 3,360, assuming departures every 30 seconds, and no delays ever. (That would also require the purchase of more capsules and larger stations, boosting the cost of the system further).
For comparison, one Metropolitan line S Stock train, introduced in 2010, can carry 1,532 people. 16 of them run every hour, moving almost 25,000 people around the capital. Even if we just look at long-distance only, a single Eurostar train carries 750 people, and five of them depart Kings Cross St Pancras in rush hour, ferrying around four thousand people through the channel tunnel.
It's the one comparison Musk simply doesn't make, and it's no small deal, either. Throughout the document are barbed references to the Californian high speed rail plan, an expensive but thoroughly developed proposal to build a conventional railway between the two cities. (Unlike the Hyperloop, the proposal actually does connect the two cities, not their outlying suburbs). We can find comparisons of cost, and of speed, but nowhere in the Hyperloop proposal is it mentioned that the current plans for the Califoria HSR give it over ten times the capacity of the Hyperloop.
Musk does note that the capacity would be "more than sufficient to transport all of the 6 million passengers traveling between Los Angeles and San Francisco areas per year", but fails to note that new transport modes increase demand. Just because it seems like the loop has enough capacity now, does not mean it won't still be over-full by the time it opens.
All the above criticisms are made even accepting the documents claims at face value. When you start to examine the claims themselves, the case gets even shakier.
Alon Levy is a transportation engineer who has taken a more thorough look at some of the major claims made by Musk. The problems he finds come in four broad areas: cost, comfort, capacity and energy consumption.
The first is the biggest. The case for Hyperloop is so strong because it can, apparently, be built for just $6bn, a tenth of the California HSR. Some of that comes from the cost cutting involved in avoiding the centre of towns and assuming land already owned by the state is free; but Levy argues that more of it is just plain wishful thinking.
For example, the cost for viaducts on California's conventional rail plan (costs which aren't presented in an "open source transportation concept", but in a far more detailed actionable plan) are around ten times what Musk plans for the Hyperloop. The capsules themselves are lighter than a train, which would reduce the cost of the viaduct somewhat, but welded steel tubes are pretty heavy themselves. A meter of the tube would weigh a little under four tonnes; as Levy points out, a meter of train weighs around half that.
Even if it is just as expensive as conventional rail, at least it's faster, right? Well, aside from the fact that door-to-door, it will likely be around the same speed unless you happen to live in a northern suburb of LA or on the east Bay, that speed causes problems. If you're going fast, you get flung out to the side at curves, and crushed down to the floor when climbing hills. Even though the Hyperloop takes account of that, it still presents a considerably more uncomfortable journey than a train: the horizontal forces are around twice that of a conventional HSR, and Levy estimates the vertical forces as between seven to ten times as high. Oh, and the 30 second headway required to fit the maximum capacity in the tubes would necessitate emergency braking twice as forceful as the US government currently allows for passengers with seat belts.
"There is no redeeming feature of Hyperloop," writes Levy. "Small things can possibly be fixed; the cost problems, the locations of the stations, and the passenger comfort issues given cost constraints can’t . . . If Musk really found a way to build viaducts for $5 million per kilometer, this is a huge thing for civil engineering in general and he should announce this in the most general context of urban transportation, rather than the niche of intercity transportation. If Musk has experiments showing that it’s possible to have sharper turns or faster deceleration than claimed by Transrapid, then he’s made a major discovery in aviation and should announce it as such. That he thinks it just applies to his project suggests he doesn’t really have any real improvement."
More than grumbles
It's easy to be a naysayer about this sort of proposal. Anything presented in a skeleton outline like this can have a myriad imaginary problems placed on it: where will the bathrooms be? Who's going to design the stations? What's the room for expansions and upgrades?
To take a draft proposal like this fairly requires an open mind. Problems will be worked out, even if cost overruns are to be expected, and shutting something down before it's even begun isn't the way to achieve progress.
But there's a difference between giving something the benefit of the doubt, and allowing unachievable promises to stand. The Hyperloop as pitched – as a $6bn project usefully linking two cities – can't exist. The problems of capacity are inherent to the plan, and even at the theoretical maximum the system will carry fewer people than a typical high speed rail network. And dumping 4,000 people an hour on the outskirts of LA will require a dramatic restructuring of the city's internal transportation networks, again not included in the cots.
If it were only a napkin scribble, the over-promising wouldn't have pernicious effects. At worst, a bit more cash might be pumped into the concept before realising its unworkable. But the whole proposal is clearly targeted at the case for the California HSR. From the off, Musk is clear, writing that "when the California 'high speed' rail was approved, I was quite disappointed, as I know many others were too." And so there is a risk that Musk will succeed in destroying the already tenuous support for a workable rail project, without the ability to offer anything else in its stead.