When Elon Musk wants something, he often does it something. The PayPal founder wants to send people to space; hence, SpaceX. He wanted to invest in cleaner automobile technology; thus a Series A investment in Tesla. Now, he wants to travel between Los Angeles and San Francisco in 30 minutes. Enter the Hyperloop.
The Hyperloop is Musk’s current project. It’s an elevated vactrain that would travel at around 600 miles per hour with top speeds closer to 800. It would run frequently between Los Angeles and the Bay Area, and it would be cheap. Musk claims construction would run to only a few billion dollars with fares at $20 for the one-way trip. Is this dreaming or is this delusional?
Since unveiling his paper on it a few days ago [pdf], reaction has ranged from incredulous to giddy. Transportation advocates are stunned by Musk’s claims — often issued with no supporting evidence — and even those with a basic level of mathematical knowledge don’t quite understand how his ideas add up. Meanwhile, lay people are awed by the idea. It’s something we’ve never seen before, and it sounds like it could bridge great gaps in short order. Plus it’s way cheaper than that whole California High Speed Rail boondoggle or so the argument goes.
As much as I like to dream big — IND Second System anyone? — color me skeptical for a variety of reasons, most of which have been expressed elsewhere. James Sinclair issued a massive takedown, and Alon Levy, for instance, calls it a loopy idea. He dispenses with a lot of Musk’s equations, questions the way this structure could withstand earthquakes and generally wants to see evidence:
There is no systematic attempt at figuring out standard practices for cost, or earthquake safety (about which the report is full of FUD about the risks of a “ground-based system”). There are no references for anything; they’re beneath the entrepreneur’s dignity. It’s fine if Musk thinks he can build certain structures for lower cost than is normal, or achieve better safety, but he should at least mention how. Instead, we get “it is expected” and “targeted” language. On Wikipedia, it would get hammered with “citation needed” and “avoid weasel words.”
…Musk’s real sin is not the elementary mistakes; it’s this lack of context. The lack of references comes from the same place, and so does the utter indifference to the unrealistically low costs. This turns it from a wrong idea that still has interesting contributions to make to a hackneyed proposal that should be dismissed and forgotten as soon as possible.
I write this not to help bury Musk; I’m not nearly famous enough to even hit a nail in his coffin. I write this to point out that, in the US, people will treat any crank seriously if he has enough money or enough prowess in another field. A sufficiently rich person is surrounded by sycophants and stenographers who won’t check his numbers against anything.
Levy isn’t the only one casting doubt on it. USA Today interviewed some scientists who raise similar concerns, and Alexis Madrigal questions the details and land acquisition process. The list of problems goes on and on and on.
In other areas, rail advocates are dismayed because Musk is one of California’s highest profile entrepreneurs, and he is essentially throwing high speed rail under the bus (or, in this case, the Hyperloop). He claims he can do a better, and since he’s a Very Important Person, Californians who are still skeptical of HSR listen. Why should we spend billions on a proven but expensive technology when we can just let Musk — who doesn’t want much more to do with the Hyperloop idea anyway — build his futuristic travel pods? Why let something actually transformative come to being when we have nifty renderings?
Dreaming big and dreaming practically in this case are two separate outcomes, but they needn’t be. There is a place for ideas like Musk’s, but there is also a place for improving the current proven modes of transit as well. We can dream up larger networks and more efficient ways to move people through areas. But one should not come at the expense of another, and we should be able to recognize something for the fantasy that it is.
Last week, Eric Jaffe wrote on The Atlantic Cities that we should stop obsessing about the next big thing. We can’t give up dreaming, but we also, Jaffe writes, cannot let it “undermine our ability to address the problems of the present…In other words, we’re far better off with good expectations than great fantasies.” The Hyperloop fantasy is a great one, but so is the world of Back to the Future II where we all have flying cars within the next 26 months. But how do we get more cars off the roads tomorrow?
I’ve seen similar analyses of Musk’s proposals, and the costs seem to leave out the last mile problem. It’s fairly cheap to build in the Central Valley, but the system still leaves a commuter with a heck of a commute to get that last bit. There’s no actual tunnel/link between the Oakland side and SFO proper. And the LA terminus isn’t in LA, but a distant suburb. The costs to go that last bit more than eat up the difference between the HSR proposal and Musk’s.
That doesn’t even touch on the engineering problems.
It’s disappointing that the proposal isn’t all that different than the vacuum tunnel proposals that were first floated decades ago (or even pneumatics that saw use in NYC before the present subway system came about).
That’s what I was thinking. We’ve been seeing this kind of stuff for as long as I can remember, and maybe the only novelty here is that the locale has shifted from New York-London or something to Kalifornia.
Do west coasters love pods because they hatch from them or what? At best, it’s a people mover combined with a vactrain.
HSR in California kind of suffers from the same problem Lhota’s park/ride plan would in New York in getting the service into locations where it actually might do some good. Land acquisition costs along the coast forced the state to relocated the line to the less-populated Central Valley, where your service threatens to become detached from the people along the route who would be most likely to use your service.
Ostensibly, if people are humming along while breaking the sound barrier inside a bunch of tubes, it wouldn’t matter time-wise if you routed them through Death Valley on an LA to SF route to save on land acquisition costs. But eventually, you’d have to route it to the big urban areas and face some of the same land acquisition and environmental concerns, let alone the other issues like earthquakes and simple liability the first time a compartment hits a bump at 800 mph. Other billionaires have their quixotic pet projects, like Jeff Bezos and his space launch efforts in the desert east of El Paso, but in those cases they’re at least going though the testing stages before they commit to the idea of shooting people up for low-orbit space vacations. Musk should probably show he can tube crash dummies from Santa Clara to San Jose first, before he gets California or the feds to start tossing tons of dollars into his Hyperloop idea.
It’s been a few years since I bothered reading up extensively on CAHSR, but can’t they just use existing local tracks for access to downtowns? That’s basically what Europeans do. HSR doesn’t do HSR speeds in the cities, it does it in the countryside.
The “short term” plan is to share tracks, yes. The “long term” plan (required by proposition) is for 100% exclusive ROW.
They can and will share some tracks with Caltrains between San Jose and SF (Downtown). On the Los Angeles end, though, there is no feasible rail rout to accommodate trans-mountain frequent train traffic. The 2 main train routes leaving from the Basin are clogged already with freight traffic, and are woefully slow to accommodate any passenger train there.
Don’t say “will” yet… there is no guarantee that project will be finished in our lifetimes…. like the full length SAS.
There is a route once trains get to Sylmar, though.
John-2, I think you are wrong on your assessment about the reasons for the Central Valley route to have been chosen.
The Central Valley IS the natural choice for any high-performance rail alignment between Bay Area and Los Angeles basin. Any other routing would add extensive tunneling. The Central Valley is flat-out, perfect for HSR construction. It is really a no-brainer.
What they decided was to serve the cities on the valley (Fresno, Merced, Bakersfield etc) instead of building a straight route closely following I-5, in the middle of nowhere, which would have been cheaper but would forego service to the cities there.
A route along the coast for CAHSR is a non-starter. The California coast is too rugged for the technical demands of high-speed rail, even if both ends of the line are by the Ocean.
But are the average incomes and travel patters of the Central Valley cities the target income demographic for HSR? Acela shares trackage with other Northeast Corridor trains, so passengers can opt for the higher-priced trains or the lower-priced regular Amtrak service (or, if they’re really frugal, go from commuter rail line to commuter rail line along the NE corridor).
HSR, or at least the dedicated section, will carry a premium price, unless it’s heavily subsidized. Can you get people in Fresno, Merced, Bakersfield, etc. to pay that cost, and do they ride the train at all if there is no regular service on the route, and it’s HSR or nothing. California has lots of high-income people, but other than a few inland pockets of wealth like Palm Springs, most of the money tends to stay within 30-40 miles of the Pacific Coast. A Central Valley HSR isn’t going to get many of those people between San Jose and Oxnard to use their service (and I’m not saying the line should be a coastal route because you’re right about the topography problems, but a line that’s not a coastal route may end up bypassing it’s potential customers).
You’re assuming HSR will look like Acela, which is the most expensive HSR system I know of per unit of distance traveled. Even the Regional is more expensive than the average Shinkansen or AVE, let alone the cheaper TGV and KTX.
I think the only HSR line yet built in the world that has a chance to be a financial bust might be an Italian one. Otherwise, I think have all covered their operating and amortization costs. This includes examples built in moderately low income countries like Spain and France. I’m pretty sure even Acela manages to cover both amortization and operating costs, and maybe even the barely slower Amtrak Northeast Regional.
The Central Valley actually has more population than the coast. Fresno and Bakersfield have metro areas of 1 million and 800k residents, respectively. The largest area on the Coast is Santa Barbara, with only 200k.
I think there’s a place for big dreamers like this, if only to get people excited about mass transportation… though if it becomes an either/or proposition it’s definately a net negative.
Looking at the rendering of the vehicle itself, the first question that comes to my mind is: Where’s the bathroom?
Youre expected to pee after you get off the theme park ride.
Musk’s big dream can be forgiven to a point because high speed rail is essentially 18th century technology pushed to the absolute limit of speed. It takes a tremendous amount of money, technology and maintenance to keep a train with 1000 people going 200+ mph down the narrow rails. Going much faster just isn’t worth the cost.
The best part of his idea is that the system is elevated. All high speed rail should mostly be elevated to eliminate potential collisions on the ground and the hard barrier a fenced route creates to wildlife and people alike.
There’s no reason an over-sized monorail type system couldn’t cruise at 300 mph through rural, flat areas such as California’s Central Valley. All the technology is already there but people don’t dream of new things the way they used to. Except maybe Musk.
Fenced berms through miles upon miles of farmland do not hurt access. Do you think there are frequent access points and roads there today?
Im quite amenable to underpasses/overpasses or pylon sections if they are needed for wildlife. The irony that has been pointed out by many people is that putting the HSR on pylons in the Central Valley is part of what drove up the cost, and them Musk comes along and ignorantly says “this whole thing will be super cheap because its on pylons!”.
This is a bug, not a feature. The reason for California HSR’s cost blowouts is that it needed to put more of the system on viaducts and in tunnels and less at-grade than originally planned.
In France, where the routes are largely at-grade, they’ve solved the barrier problem in two ways. First, they negotiate land swap deals with the farmers that leave their plots all on one side of the tracks, so there’s no need to cross. Second, they build a few grade separations, which are cheap in rural areas, including some crossings for wildlife.
In urban areas it’s usually impossible to do this at-grade, but most of the system is not in urban areas.
I think the concern about the pylons is misplaced. Elevated roadways aren’t expensive because of the pylons but because of the viaduct. If they can prefabricate the tube sections and join them on site via orbital welding then the construction costs could be significantly lower. Orbital welding is a mechanized process. Musk is essentially saying that unlike a normal viaduct, this one can be built using robots.
Viaduct sections can be prefabricated too, and then simply lifted into place on top of pillars. Other than building the foundations which has to happen either way, this process is already largely mechanized. There’s no reason to believe that vacuum tubes would be significantly cheaper, and certainly not by a factor of 10.
Flat decks are not self supporting, they require something to serve as a deck truss. So even if sections are prefabricated in one piece, their construction, transport, lifting, and attachment is more intricate and therefore expensive. They’re likely to weigh more for a given amount of load carrying capacity and are likely to use a greater number of parts and steps even in the prefabrication process. And they’re likely to require lots of manual welds in hard to reach places in order to join any two sections together.
Tubes are not self-supporting either, particularly when they need to be strong enough to withstand an ambient-vs-vacuum pressure differential. Even if you do save some weight, it’s not going to be the factor-of-10 cost difference that Musk proposes.
If you don’t like roads and railroads as a base of comparison, then look at the inflation-adjusted cost of the Alaska pipeline, and then adjust for the fact that Hyperloop is intended to be twice as thick, 2.5 times as wide, and far more precise in its tolerance standards.
Okay, so checking out the Wikipedia page I see that the Alaska pipeline was built on inhospitable terrain in extreme temperatures, required new highways, tens of thousands of workers living in camps and putting in a lot of manual labor. The average worker apparently got paid more than a US senator. All welding was done manually, required extensive certification, and a lot of the welds were still defective. The pipe was just regular steel tubes that required extensive pylons for support. I suppose if Musk’s project was to be built the same way, the costs would be astronomical.
If Musk’s idea is built the same way, the system will explode as soon as someone tries to pump the air out of it. The level of precision required is aerospace-grade, not pipeline- or rail-grade. It requires state of the art machinery, state of the art engineering, safety-critical fabrication (ideally without the part where the O-rings get brittle). I’m sort of willing to believe that Musk has the technological capability to get it built, but not that he can build it for less than ordinary viaducts.
That’s exactly what orbital welders are for. Developed by the aerospace industry for precision welds on tubing, they are also far cheaper than what the Alaska pipeline had spent on workers. It’s really not that hard or expensive to create an airtight seal with reinforced factory-produced steel tubes that are welded together automatically. You line the tube with an interlocking concrete support structure and weld the steel shut; essentially it becomes a steel-lined concrete sewer pipe that can span far greater lengths between pylons than the simple steel tubes that’s used for oil pipelines.
I’m glad to see well-reasoned critique of this concept. I’m all for futuristic proposals, but don’t present them as something they’re not. In what universe would anyone think that this could be cheaper than widely available rail technology? One of the supposed advantages is the ability to build on concrete pylons, which is already done with rail! The proposal completely ignores legal/NIMBY expenses, station infrastructure, safety certification, etc. I love a far-out concept, but don’t make unfounded claims comparing it to dependable, proven high-speed rail.
I’m glad to see well-reasoned critique of this concept. I’m all for futuristic proposals, but don’t present them as something they’re not. In what universe would anyone think that this could be cheaper than widely available rail technology? One of the supposed advantages is the ability to build on concrete pylons, which is already done with rail! The proposal completely ignores legal/NIMBY expenses, station infrastructure, safety (how do you get out if it breaks down?), etc. The pods look incredibly uncomfortable! I love a far-out concept, but don’t make unfounded claims comparing it to dependable, proven high-speed rail.
Glad to see ole Elon’s been catching up on old episodes of “Extreme Engineering”
I wonder; isn’t MagLev now a proven technology at this point? I believe it is in daily use to & from the Shanghai Airport. Why isn’t that being considered for California’s project if it has to be all new construction anyway? Is it more costly to operate?
Significantly more. China originally planned to build Maglev everywhere. They stuck to traditional rail due to cost. And thats China, where cost isnt an issue.
A bit more cynically, it’s because there’s a lot more profit to be had in stealing HSR technology than stealing Transrapid technology. After all, you can’t exactly sell something when you’re the only one who can afford it.
It seems to me that the main concerns being expressed here are that of costs. That’s fair, but to be the devil’s advocate, Elon Musk has a history of finding cost effective solutions to existing problems, from money to car manufacture to rocket engine design. Why Alon Levy considers this a bad think puzzles me, especially considering that one of the biggest recurring concerns around here is the purportedly exorbitant costs of developing mass transit the traditional way in the US. Musk’s various successes to date were never based on throwing some money at a few off the shelf parts in order to make something work and it wasn’t based on blindly reinventing the wheel but to take an honest look at the actual shortfalls of existing solutions. In that light, criticism of his proposal should take into account more than just existing cost structures and assuming that his costs would be the same.
yes he’s an innovator… but Tesla’s are not cheap cars at all… and still not yet guaranteed to survive.
Teslas’ cost less than a comparable car built using traditional materials and manufacturing. They can’t dictate battery costs, but those are coming down. They’re turning a profit while major car manufacturers are surviving on bailouts.
Tesla’s not actually profitable according to generally accepted accounting principles.
And when someone with zero history of civil engineering tells me he can build an elevated system at $5 million per kilometer without citations, I reserve the right to call bullshit.
sorry – but the auto industry is not that easy. it will take 10-20 years before we know if they will have staying power.
Is there an added value for a company to be around 10-20 years?
It’s partially because the most expensive parts of CAHSR (the inner city segments) would also be ridiculously expensive with a tube (which can’t exactly be elevated through Oakland and SF). If you don’t do it at all, then you end up wasting more time trying to get to the terminals than you would be spending if a slower HSR train made several intermediate stops in the metropolitan areas.
Is it that bad to dream? It takes 30-50 years to build 1 subway line to completion and at what cost, 3 – 8 to convert a route to incorporate +SBS+ treatment. The GOP attempts to pull infrastructure budget, especially urban rail $ every chance they get. And there appears to be no resolution of any of this in sight.
In this reality, dreaming sounds pretty good to me…
When the numbers are fantasy, yes, it’s bad to dream.
Stay tuned for Hyperloop 2: The Segway from Al Capone’s Vault.
At best this is a poorly thought out pipe dream. At worst, a feeble and tardy attempt to sow FUD over the California HSR project.
if musk wants to take a stab at this on his own private land using his own private dollars more power to him. maybe he will surprise all again. until that time maybe he could suggest ways to lower cost of HSR in california since he’s a super billionaire genius and all.
Seriously, this proposal is a joke. Even Musk himself is not interested in investing in this Hyperloop. If anything, I perceive this to be more of an attempt to derail the HSR project which clearly has its own issues. Everything Musk has done has been incremental. For example, the electric car — cars, electric motors and batteries existed before Musk, what he did which others failed to do was to put them together in a manner that allowed for a viable product. The story with Ebay is the same — payment systems existed before Ebay, but the incremental contribution of making it peer to peer was really good, no questions asked. SpaceX fits a similar concept — rockets have existed for a long time, but he and his team figured out how to do it on a budget much smaller than what the government was spending. The conclusion is that Musk is good at taking existing technology and making it better.
This Hyperloop is different. To the best of my knowledge, there is no working system in the world transporting people or any form of cargo in partially evacuated tubes longer than 50 yards (that number is arbitrary, but the only system I know of is the stuff used in drive-through banking and that is nowhere close to even 50 yards). This is not about taking existing technology and making it cheaper and better, this is about inventing the technology and Musk has no track record doing so. The engineering problems are so numerous and daunting that this has no chance of happening even if someone invested a trillion dollars in it.
Even if it were technologically sound and the costs made sense, the ridiculous capacity constraints make it not worth investing billions in. 3k people per hour is small peanuts compared to a HSR that can seat 1000 on a single train easily.
It’s odd to think there is no better idea than metal wheels on two rails, but maybe it’s the case. My question would be, is this idea so ridiculous, even on paper, that it’s not even worth a demonstration?