(2 pm. – promoted by ek hornbeck)
Cross-posted from its origin station at Voices on the Square
On Monday, entrepreneur Elon Musk launched on attack on the California HSR system in the guise of a pie in the sky alternative that he has dubbed the “Hyperloop”. Now, I got into this topic from the back end, since I waited for the technical people to download the PDF and chew into it before giving it a serious look, so when I first encountered the notion floated that this is just a car builder (well, an electric car builder) attacking a rival form of transport, I thought that might involve some shaky inference regarding motive for otherwise puzzling statements …
… but then I read the first paragraph of the blog post where he introduced the proposal, and there really isn’t any doubt:
When the California “high speed” rail was approved, I was quite disappointed, as I know many others were too. How could it be that the home of Silicon Valley and JPL – doing incredible things like indexing all the world’s knowledge and putting rovers on Mars – would build a bullet train that is both one of the most expensive per mile and one of the slowest in the world? Note, I am hedging my statement slightly by saying “one of”. The head of the California high speed rail project called me to complain that it wasn’t the very slowest bullet train nor the very most expensive per mile.
So this is explicitly a proposal from the guy who made big bucks on an internet payment system, Paypal, showing how the California HSR is old, outdated technology and if he wasn’t busy doing other things, why, he could give us an intercity transport system that would knock our socks off.
Earning the Title Hype-Loop
But then, given the same internet on which Elon Musk made his bucks, people downloaded the proposal and tore it to shreds. Basically, the cost estimates are hype, the passenger capacity estimates are hype, the actual connection of LA to SF is hype, the passenger comfort claims are hype … indeed, there are claims in the piece about the California HSR system that are outright lies. There is every reason to dub it the Hype-Loop.
Alon Levy, mathematician who blogs about transport issues on the side:
My specific problems are that Hyperloop a) made up the cost projections, b) has awful passenger comfort, c) has very little capacity, and d) lies about energy consumption of conventional HSR. All of these come from Musk’s complex in which he must reinvent everything and ignore prior work done in the field; these also raise doubts about the systems safety that he claims is impeccable.
What is this awful passenger comfort:
… This is worse than sideways acceleration: track standards for vertical acceleration are tighter than for horizontal acceleration, about 0.5-0.67 m/s^2, one tenth to one seventh what Musk wants to subject his passengers to. It’s not transportation; it’s a barf ride.
Stop And Move, from Fresno-based urban policy, development and transportation blogger James Sinclair:
Amusingly enough, the California HSR budget for the Central Valley is under $10 billion. Ie, in the same ball-park as this proposal. The reason the HSR project is going to cost $60 billion is because it has to face an uncomfortable truth; actually getting to LA and SF is expensive. Very expensive. That’s where there’s no free land. That’s where you have years of property acquisition. In the shorter term, the plan for HSR is to simply share existing tracks, which the Hyperloop can’t do.
So either the budget explodes, or the project doesn’t actually serve the main cities. You can’t have it both ways.
In the comments of the California HSR blog, Clem, who runs the Caltrain HSR Compatibility Blog, has a list of eight “show stoppers, including security, which addresses the claim that there will be no need to fence this system off, since the pylons and the tub itself provides all the security required:
Show Stopper #3: Security-wise, it’s easy to destroy this system by making a small dent in the tube. Yes, it’s one inch thick steel, but there are easily attainable ways to dent one-inch steel. When you hit a small bump protruding inside the tube while going at Mach 0.9, nothing good can possibly happen regardless of capsule suspension design. As proposed the system is impossible to secure.
… and passenger capacity, based on egregiously unrealistic headways:
Show Stopper #1: There will always be the risk of equipment failure or tube damage that will cause a sudden and unplanned stop, which dictates the headway between vehicles. This minimum headway, regardless of how fancy and automated the traffic control system, is a simple and necessary consequence of “shit happens” and creates an inherent capacity limitation for fixed-guideway transportation systems, of which the Hyperloop is just another example, after you look past all the high tech.
The minimum headway consists of the following time contributions:
- the time for the preceding capsule to clear a junction and free the routing through the junction to be changed for the following capsule
- the time for the route through the junction to be changed
- the time for the control system to confirm the new route is clear
- the time for the following capsule to perform an emergency stop if the new route is not clear for whatever reason
- a bit of extra time for safety margin.
This is no different than for HSR, and the resulting minimum headways are usually several times longer than the emergency braking distance. HSR deals with the capacity limitation imposed by the minimum headway by (1) packing a huge number of people into one vehicle, e.g. two double-decker trains coupled together with over 1200 passengers, and (2) by operating at top speeds that strike an optimal balance between reasonable braking distances and close headways. Hyperloop can’t strike this balance: in order to work, it requires very high speeds (and greater emergency braking distances), can carry only a few passengers per capsule, and requires relatively short headways. The two-minute headway assumed in section 4.1 is speculative at best, and it’s not at all a given that the Hyperloop could ever provide the same transportation throughput (in passengers per hour per direction) than plain old HSR. Not by a long shot. The minimum headway question is fixed guideway system design 101, and may be the Achilles’ heel of the whole concept. It is not sufficiently addressed in the document.
The project has a billed cost of $6b to get from LA to SF, but that would actually be Sylmar toward the northern edge of LA County through to the southern edge of Oakland in the SF-Oakland metropolitan area. Getting it into downtown SF and downtown LA would cost dramatically more.
The project has a billed maximum capacity of 3,360 passengers each hour, 20%-25% of the maximum capacity of the HSR and about 34% more capacity than a lane of freeway at an average 1.25 passengers per vehicle. But as Greater Greater Washington points out, when the egregiously unrealistic headways are modified to merely optimistic headways, the capacity of the system falls away:
That means that the minimum separation between pods is probably closer to 80 seconds or more. Not a big deal. It still means 45 departures per hour. But that’s only 1,260 passengers per hour in capacity. That’s 10% of what the California High-Speed Rail can carry.
With a capacity of 1,260 passengers per tube, that means that the Hyperloop would need 10 tubes in each direction (not 1) to move the same number of passengers as the proposed high-speed line. And that would push the cost up by 10, which is actually more than the cost of the HSR.
So the project has a billed cost of $6b to get from LA to SF, because its built on pylons, so land acquisition is not required, making it cheaper. When we adjust for the promised capacity, we would have to build four to five systems to match the maximum HSR system capacity, so the headline “$6b cost” is not really a 90% discount: its actually a 40% to 60% cost discount. If we adjust it for a still optimistic but less unrealistic headway, the discount disappears.
And there’s no reason to believe that building on pylons would in reality end up to be cheaper, as Alon Levy points out:
In reality, an all-elevated system is a bug rather than a feature. Central Valley land is cheap; pylons are expensive, as can be readily seen by the costs of elevated highways and trains all over the world. The unit costs for viaducts on California HSR, without overhead and management fees, are already several times as high as Musk’s cost: as per PDF-page 15 of the cost overrun breakdown, unit costs for viaducts range from $50 million to $80 million per mile. Overheads and contingencies convert per-mile cost almost perfectly to per-km costs. And yet Musk thinks he can build more than 500 km of viaduct for $2.5 billion, as per PDF-page 28 of his proposal: a tenth the unit cost. The unrealistically low tunnel unit cost is at least excused on PDF-page 31 on the grounds that the tunnel diameter is low (this can also be done with trains if they’re as narrow as Hyperloop, whose capsule seating is 2-abreast rather than 4- or 5-abreast as on HSR; see below on capacity). The low viaduct unit cost is not.
Solving the Wrong Problem Normally Doesn’t Solve the Problem
What is the problem that Elon Musk is solving? The claim is that its that the California HSR system is (a) slow and (b) expensive.
What about the slow? LA and San Francisco are 341 miles apart (as the crow flies), and at 2hrs, 40 minutes, that means that the average line-of-sight speed of the express California HSR is 130mph. The trains may be able to go 220mph, but given the terrain, if an HSR system was built on a straight line route, the cost would be in excess of $100b. It is necessary to run through the Central Valley where the HSR alignment is relatively cheap. But on the other hand, this means that the California HSR corridor is able to serve Central Valley communities, where HSR service is worth more per person than in LA or San Francisco, both because the Central Valley is underserved with intercity transport at present, and because it is closer to LA and San Francisco than either are to each other.
Then, in an alignment choice that is controversial in some quarters, the southern access to the Central Valley is via the more easterly Tehachapi Pass rather than the more direct western Tejon Pass. Whether this is because of the greater difficulty of the Tejon Pass alignment or the political efforts of Palmdale can be the cause of endless online discussion involving a small number of dedicated Tejon Pass alignment partisans, but the end result is to take a semi-circle from Sylmar in northern LA county through the Tehachapi Pass instead of a nearly direct shot north from Sylmar to Bakersfield.
But its quite clear that Elon Musk doesn’t really give a damn about door to door transit speed, if he is anchoring his Hyperloop in Sylmar and not bothering to get to downtown LA. So the transit speed is not the issue, its that the steel wheel on steel rail HSR will “only” goes 220mph when it hits the flat, straight alignments of the Central Valley, and 220mph just isn’t as sexy as 700mph.
And cost? That is comparing apples to oranges, since to get the same capacity on any realistic headway, his system would cost as much as the Orange County through Downtown LA to Fresno to San Jose to Downtown SF California HSR, and then only by assuming magical pylon construction costs, and even then would only run from northern LA county to the southern edges of Oakland
And of course the capacity of his proposed system drops if any appreciable share of passengers are overweight or have long stays on the other side, since he assumes 220lbs per passenger including luggage.
And of course, his assumed sideways accelerations seem likely to make the last part of the half hour quite unpleasant. With half an hour in a sardine can, and an hour or more transit on either side, somebody in a pod of 20+ people is likely to have eaten before entering the sardine can, so it will be a common occurrence for someone to barf, which will likely cause someone else just holding it in to barf.
All of this, of course, starting a decade or more in the future, since even if the project was funded today, it would still first be an R&D project that will have to prove out Elon Musk’s extraordinarily optimistic assumptions before the project can go ahead.
Of course, Elon Musk is just putting this proposal out there: he is not going to be pursuing it himself. He’s busy building expansive electric cars for rich people and working on the SpaceX project, so doesn’t have time to take his proposal from concept PDF through R&D to implementation.
What is the Problem to be Solved?
The fact that Elon Musk was able to make a pile of money being the front-runner on getting substantial market share for an online payments system up doesn’t make him any less of a dilettante when it comes to transport technology. The problem to be solve is:
- Provide an alternative intercity transport system for California;
- 1. Connecting all of the largest and most of the intermediate population centers
- 2. That can be powered by sustainable, renewable power;
- 3. Using already-existing transport technology
- 4. That can attract a substantial share of the existing intercity transport demand.
The California HSR system meets criteria 2-4 by design. Unlike the promised half hour roller coaster ride in an enclosed sardine can, the HSR would offer an opportunity to get work done or watch a movie, get up and stretch your legs, and to go get a snack if you didn’t have a chance to grab on before catching the train. Over Phases 1 and 2, the California HSR system connects all of the largest and most of the intermediate population centers in the state.
By contrast, the Hyperloop only meets criteria 2. It is proposed to use an untested, yet to be developed technology, and as proposed would never offer trip comfort that would attract a substantial share of passengers, so as proposed, it fails criteria 3 and 4.
Regarding the number one target for the California HSR system, consider the top fifteen California urban areas (2010 census), which are the ones with populations greater than 350,000 ~ 15 of about 110 urban areas across the US. These are not Metropolitan Areas, but continuous urban areas with boundaries defined by population density. These are by US Rank, Name and population.
Here are the ones served by both the California HSR and the Hyperloop (HSR1/HSR2 is HSR Phase 1 and Phase1, Hyperloop1/Hyperloop2 is the base Hyperloop proposals and the branch network system:
- (#2) Los Angeles–Long Beach–Anaheim, CA, 12,150,996, HSR1, Hyperloop1 (just barely)
- (#13) San Francisco–Oakland, CA, 3,281,212, HSR1, Hyperloop1
- (#15) San Diego, CA, 2,956,746, HSR2, Hyperloop2
- (#28) Sacramento, CA, 1,723,634, HSR2, Hyperloop2
- (#63) Fresno, CA, 654,628, HSR1, Hyperloop2
Here are the ones directly served by the HSR system, but not the Hyperloop System:
- (#22) Riverside – San Bernandino, 1,932,666, HSR2
- (#29) San Jose, CA, 1,664,496, HSR1
- (#79) Bakersfield, CA, 523,994, HSR1
- (#87) Murrieta–Temecula–Menifee, CA, 441,546, HSR2
- (#102) Stockton, CA, 370,583, HSR2
- (#105) Modesto, CA, 358,172, HSR2
- (#112) Lancaster–Palmdale, CA, 341,219, HSR1
Here are the ones served by the Hyperloop but not the HSR:
And here are the ones directly served by neither, with their Amtrak California service:
- (#66) Concord, CA, 615,968 (Capital Corridor)
- (#69) Mission Viejo–Lake Forest–San Clemente, CA, 583,681 (Surfliner)
- (#103) Oxnard, CA, 367,260, (Surfliner)
The two cities along the Surfliner route are offered superior cross-platform connections to the Phase 1 California HSR system at Anaheim and LA Union Station, while if the Hyperloop actually ends on the southern edge of Oakland, then the city along the Capital Corridor route is offered a superior connection by the Hyperloop.
All in all, the California HSR plan not only offers superior intercity transportation access in downtown Los Angeles and San Francisco, but also offers effective intercity transportation services far deeper into the California urban hierarchy than the Hyperloop proposal.
Now, the California HSR system certainly is not exactly the project as I would design it, but then again the Hyperloop would not be as Elon Musk would design it, if it passed through R&D and turned out to have one or more design alternatives that met some transport task … at that point it would go into the political meat grinder, backed up by existing private and public property rights. Whether pursued as a public project or a private one, there would be interests objecting to it, interests supporting it in general but on a different alignment, and, particularly, since it offer an airplane-style point to point service rather than a train-style route corridor service, interests arguing that resources invested in serving the point to point corridor chosen should rather be spent providing some other service to some other part of the state.
It is, after all, the politicial meat grinder backed up by existing private and public property rights that is responsible for much of the cost of the California HSR system. Put the system in the hands of a “benevolant dictator” whose dictate cannot be over-ridden and who, just conveniently (and slightly implausibly), wants to put it through in the most cost effective way available, and tens of billions could be trimmed off the project cost.
Which is what this proposal highlights: the difference between imagining a major intercity transport system and actually working through the plan to implement it would quite easily account for the promised 40%-60% cost discount per passenger capacity, even if there was a shred of credibility to that cost estimate.
Now, I’m all in favor of pursuing research and development on this technology as a transport option. However, it shouldn’t be pursued on the basis of implausible hype. Rather than offering a cheaper and faster intercity transport option, this seems most likely to offer a more expensive and faster intercity transport option. And if this technology is ever developed, whether its worth the extra cost to go however much faster this really permits us to go will be the debate.
Indeed, despite Elon Musk seeming to think this offers an alternative to an HSR system, the most cost-effective place for one or both ends of a Hyperloop Tube corridor could well be at an HSR station, allowing passengers to be effectively recruited and distributed over a much greater area around the hyperloop terminus without requiring massive acreage to be set aside for parking lots.
Conversations, Considerations and Contemplations
As always, rather looking for some overarching conclusion, I now open the floor to the comments of those reading.
If you have an issue on some other area of sustainable transport or sustainable energy production, please feel free to start a new main comment. To avoid confusion among those who might be tempted to yell “off topic!”, feel free to use the shorthand “NT:” in the subject line when introducing this kind of new topic.
And if you have a topic in sustainable transport or energy that you want me to take a look at in the coming month, be sure to include that as well.