You can’t prevent temperature changes to the tube - Even your huge air conditioner in a ring around an ‘inner’ tube wouldn’t do it. For one thing, your ‘outer’ tube is now expanding and contracting. For another, it would take a colossal amount of power to maintain temperature stability of a huge metal mass 350 km long.
I wonder if anyone’s done the math to calculate the temperature change to the tube as a result of pumping the air out? Adiabatic expansion will definitely lower the temperature. It seems to me that every depressurization cycle is going to cause a rapid drop in temperature, followed by an increase in temperature as the system re-heats. That might add a lot of stress.
Also, when calculating the expansion of the tube, you can’t be considering just air temperature. Earlier I mentioned 120 degrees as a reasonable ballpark number for temperature change, and several people disputed that it would be that much in the California area. But we’re talking about surface temperature here. What’s the surface temperature of steel with the sun beating down on it? Maybe that’s why Musk wants to put the tube in the shade by putting solar panels over it. If the sun hits it directly, then depending on its color the temperature could go well over 120 degrees - it could get hot enough to burn you, just like a car body in the sun.
Is it polished? Is it staying polished?
The dual-walled design would greatly reduce the effect of solar heating on the carrier tube and help maintain a consistent temperature along its length. I would be concerned about localized temperature differentials, such as a storm lowering the temperature of one section – what happens to the structural integrity of steel when it experiences frequent localized temperature differences over time?
Seriously, I think the evacuated tube concept is asking for trouble. It would make almost as much sense to just pump air through it, drastically reducing wind resistance for the cars and greatly reducing the depressurization risk for the passengers. Of course, the problem with that is that you would have to filter the inflow to reduce dust-scouring, and California is an especially dusty place.
Pumping the air (at 700 mph?) through a 350-mile tube is not feasible.
Maybe that’s why the suggested route is SF to LA? 6-8 hours drive time, depending on how nuts you are, with minimal lead-in time (no security, no parking at the airport, no renting a car at the other time); it’s all travel time. Granted that the system can only handle people in the tens of thousands per day, I could see high-end consultants, executives & entertainment industry people packing this thing to the rafters.
I would not suggest it would be (700 mph would be shockwave speed, that would be prohibitive), but there might be some way to stage the pumping. If one could somehow control the scouring issue, the tube itself could be perforated for inflow (this would address the claustrophobia question as well. A one or two hundred mph wind would allow 700 mph vehicles to travel subsonic at sea level. And staged airflow might offer a workable way to deal with possible track switching (by varying tube pressures).
Not ideal by a long-shot, not if you want really high speeds, but remotely feasible (which is kind of how I see hyperloop).
The problem with this idea has nothing to do with shockwaves, but with the frictional losses of the air passing the inside surface of the tube and from turbulence and swirl of the air stream. It’s been a long time since I had to do any calculations of fluid dynamics in tubes (in college, almost 30 years ago), but I do know this would be a huge problem. If you look at the ratio of distance to the diameter of the pipe, the pipe suddenly looks like an incredibly small, very long straw. There’s a fixed limit to how much air you’re going to be able to pump through that.
The efficiency of the pumping also goes down with every bend in the tube, and anything inside the tube like wiring or the track itself, or the bellows and interfaces of expansion joints would cost energy and contribute to turbulence.
I did notice that the various semi-plausible fixes to make the loop possible involve adding a lot more weight, mass, space, or some combination of the above would make it utterly impossible to fit this within the space intended, or would require a lot more bracing and land. Of course, this would then tend to ruin Musk’s already-dodgy assumptions about required land costs, which even at a base are probably an order of magnitude too low. I’m also not at all sure that you could seriously fit the loop within the I-5 median unless everything works exactly the way presented in the abstract.
The other problem with the tube concept is that it doesn’t really lend itself to a network, the same basic problem that really hurts passenger trains - they only make sense for people living close to major node points. Further, once you build it, you’ve got a huge sunk cost, and if people don’t do what you expect, you’ve just wasted a massive amount of money.
A friend emailed the PDF to me a few days ago, and from where I sit it’s inventive but pretty ridiculous to implement. Musk has a lot of effort in on the engineering side and not enough on the economics…
Suppose for the sake of discussion that all the engineering issues were solved and the proposed line were “poofed” into existence. Suppose further that similar lines were created all over the US so that we have a network of hyperloops connecting all the big cities. Just imagine it - it only takes a moment…
What happens that makes our lives better, really? Not much, the small number of passengers that this can carry is about the same as a lane of interstate (1 person every 4 seconds). And as our meager intercity traffic seems to suggest, there’s really not that much value in whisking individuals from city X to city Y - most everything they need is probably just as easily acquired in the city they left behind.
With communication infratructure constantly getting better (telecommuting being just one example, online catalogs and bills instead of mail, etc.) and the country becoming homogenized to the point where most every city is a near clone of the next, it’s kind of a quaint 20th century idea to me that the thing we really need is for trains to go faster? It’s just not as relevant anymore…
For the select few that really need to go there in person fast, air travel already fills the niche well and with considerably more flexibility.
For comparison, consider a dedicated pair of airports from LA to SF with facing runways (no long banking approaches or tedious taxiways) and a stable of 737’s compared against Musk’s proposal. One plane every 15 minutes gives similar throughput, similar speed, similar convenience, and for a considerably more tolerable economic outlook since they can better carry freight. As well as boasting a proven safety record and solved engineering problems.
And you still need to rent a car or call a cab when you get there, which is a big part of the reason people often drive these medium hops rather than fly. When all the costs are considered, even a $20 plane ticket isn’t justifiable in many cases if you can put 5 people in the SUV and drive straight to grandma’s back in the old neighborhood.
Now, the problem that really needs solving is intra-city travel - there’s a good economic case to be made there. Shaving a few minutes off for the grand majority stuck in traffic downtown multiplied across the thousands of commuters makes a lot more sense than doing the same for the small minority travelling city-to-city. Given that doing either one has an enormous cost associated with it.
The Hyperloop has a throughput of 28 people every 30 s, or 56 people/min. The largest 737 carries 215 people, which at every 15 min has a throughput of 14 people/min.
But I’ll admit, it *would *be pretty exciting to have an airport where incoming and outgoing aircraft are on the same compass direction.
And kind of problematic for half the traffic. No one wants to take off with a tailwind, no one wants to land with a tailwind, and crosswinds, well, not much fun. Banking to or from the wind is a very small price to pay for control. No, what would be exciting would be an airport build like a great big bowl, where the jets bank into the runway and spiral down to the terminal.
Well it’s interesting that you would say that since Musk has a degree in economics. It bears repetition, but the begining of the PDF clearly lays out the economic advantage of the particular route LA to SF. It’s at an intermediate length distance where it’s too long for convenient car travel and too short for air travel. It has its niche, and that’s precicely what is being exploited here. The wrong assumption people make us that its going to be built out in the Midwest or something… :rolleyes: or all over the world :dubious:
Dr Strangelove, I would hope it was obvious but I used Musk’s 2m interval (as suggested by the 1 person per 4 seconds remark) and a not-quite-full 210 passenger 737 for the calculation. Feel free to swap in different numbers, but it doesn’t change the outcome much as shown below…
Back to economics…
The current air traffic between SF and LA (which travels at similar speeds to the Hyperloop already) is a fraction of the throughput this tube gadget is sized for, so it’s probably safe to assume it’s not going to be chock full of passengers on every available train. As pointed out by Exapno Mapcase earlier in this thread.
In fact, the entire population of both cities has to take 3 Hyperloop rides a year to keep it full. Which is required to break even in 20 years, maybe, if there are no cost overruns at all…(per the PDF)? And a single empty seat per car is a 3% revenue hit?
(hyperloop capacity: 14.7 million rides a year using full cars every 2m in both directions; pop of SF and LA combined is 4.6 million, dividing gives 3.2 trips per person per year)
Assuming 100% uniform usage and performance for any capital equipment investment is about the same as assuming “no friction, we’ll just have perpetual motion instead” in engineering. There’s always bottlenecks, downtime, and irregular demand. What happens at Christmas when everyone travels?
What happens on election night when everyone is home voting (or glued to the set watching the outcome?) If there’s never an empty seat, there’s no way to just show up and buy a ride on a whim; some percentage of extra capacity needs to be designed in or the system can’t work. For example, at a bus stop you can wait in confidence knowing you’ll be almost always be able to get on the next bus that arrives. What if the stop (or bus) is so crowded that you might have to wait for a large number of busses before you can get on instead?
Is the demand for travel perfectly symmetric between both cities or will there be mostly full cars one direction and mostly empties coming back at certain times?
All this stuff cranks up the operating costs and pushes back the payback period, so it seems overwhelmingly likely to be a bad idea to build this.
A crucial point to keep in mind is that cities aren’t forever either - they bust and boom in cycles just like the economy entire (albeit generally more slowly). So what if one of them “pulls a Detroit?” and falls into economic hardship? Or some other city grows at a few % more so in 30 years it’s double the size and now a more attractive destination? You can always fly somewhere else once you’re in the air, but a railway is nigh on immoveable once built. Dead rail lines are all over the place.
Back to my main question, why would the public invest in this expensive infrastructure that saves a few minutes for a small fraction of the city travelers (say 5% for the sake of discussion - so one person in 20 uses the hyperloop routinely) rather than a similar investment in intra-city improvements that benefit the other 95%? Do LA dwellers say to themselves “Man, I really wish I was in SF right NOW!” often enough to justify this? And how much would they pay for it?
Is it merely because it’s “cool?” Seems like it’s a better idea to “beef up BART” instead if there’s any real concern for the plight of the everday commuter or for encouraging the economic growth of the cities…
Since it’s not likely to pay for itself through the farebox it’s going to cost somebody quite handsomely. It’s not like the Transcontinental Railroad days where there were no good options at all for crossing the area until the train was built and it was the only game in town.
I don’t care what kind of degree he has, building this in California is going to ridiculously expensive.
In the Midwest, it might well be plausible. IMHO. IANAEconomist. (I am an engineer, but not this kind)
Dig a tunnel, shore it with a carbon fibre pipe, run a chain its entire length with hooks every quarter mile, install a stationary steam engine at one end and voila.
It could double as a rocket wash.
Well it helps if you’re up to speed on the very basics, so I’ll just mention it to make sure: in 2008 California voters passed a proposition to fund and build a high speed train. It turns out that due to various technical reasons the actual implementation of it would turn out a fairly expensive and slow model. Hence Elon Musk introduced an alternate concept whereas it would be significantly faster and more affordable.
As to the rest of your diatribe of cities not lasting forever is… Well… Sad. :rolleyes: With that kind of attitude nothing would ever get planned or built.
And why is that so? First of all, as on any project components come from many places. Nothing is ever “entirely” built in one place. Steel mills would have to churn out similar levels of plates to create the tubes regardless of its final destination is California or Nebraska. As a side benefit, California is already the major power in the aeronautics and aerospace industry, so people with that particular design and manufacturing expertise are already here.
Plus, if the people of the Midwest want a hyperloop, then nothing is stopping them from getting one. They just need to pass their own propositions to fund it, or vote for the right state legislature representatives to draw up laws to fund it as required in each state. What’s in discussion now is the fact that California is committed to raising and spending money to build a high speed train. Also, there’s a reason why “crazy” ideas like this spring out of California, and not some regressive Podunk Midwest state 
Could you share that reason with us?
I agree, but your comments don’t really address the economics in depth.
You do have it right that there are two fundamentals of transports, intra-city and inter-city. Both are crucial, and each is very different. Teleportation would solve both but short of that you have to break the problem down into nodes of departure and arrival. The larger and fewer the nodes, the better mass transport works. The smaller and more numerous the nodes, the better individual transport works.
California’s population is currently 38 million, more than all of Canada’s. It has two exceptionally large nodes in L.A. and S.F., along with other large nodes in San Diego and Sacramento. Speeding connections between nodes by an order of magnitude is the difference between traveling by horse and traveling by car. It’s huge, and would change everyday habits and the economy, just as the auto did. California, with its large population and relatively small area, is therefore the prime place to locate this. It would be similar to the effect of bullet trains in Japan. The economy would reorient around it.
My guess is that’s one reason why Musk made his passenger estimates so high. Obviously that drives payback time down, but a 15,000,000 volume for the primary route, plus millions more for the secondary routes, becomes an economic driver in its own right. Get the price low enough for five day back and forth travel, comparable to a week’s gas, say, and the local economy can expend into new territories. Air freight exists alongside truck freight and railroad freight: each has a niche and we depend on all three (along with ocean shipping) in an economy they made national and then global.
Freight is a more important future than passengers, in many ways. You can’t email freight. People will continue to move freely daily, though. We are not going to sit in our homes and do everything via computer. Traffic has worsened in the Internet age. Highways are overloaded and airlines are overloaded.
Intra-city transport is at standstill. People will not give up the incredible freedom of having their own infinitely flexible pod. Cities are built around the automobile, and that includes every city in North America including metro New York. We’ll need a similar 100 years of de-evolution to change that and there is no will to do so.
Only inter-city nodal transport has a chance of change. Musk wants to do that in mass at a low cost. That’s the ideal, but none of his numbers make sense. Lower numbers and higher costs will not replace current systems and will never even get built. That’s also the problem with projected high-speed trains. They make sense only in a tiny few corridors and that’s only if their costs - both to build and to operate and to ride - would be much lower than expected. The country has too many nodes that are too diffuse and too far apart to be successfully served by any single mode of transport. Worse, each competing one - cars and trains and planes - pulls needed passengers away from the others, making them all weaker.
It’s an insoluble problem today. A breakthrough is needed. A cheap superspeed mass passenger, auto, and freight carrier would indeed be ideal. I don’t believe that the hyperloop is either mass enough or cheap enough, and having a fraction of either would doom it to failure.