I don’t think the pertinent question is “How feasible is Musk’s plan?” but “Is Musk’s plan more feasible than the proposed solution?”
Already did. California is home to the highest number and concentration of aerospace engineers anywhere.
Musk’s plan may be closer to feasible but still not feasible. Therefore its feasibility is the pertinent question.
What’s unfeasible about it? And why do you think his numbers are not correct? What would be correct numbers for this project?
Not to be a Junior Mod or Exapno Mapcase’s defender (he don’t need one, trust me), but read the damn thread. The whole thing is an explanation of why his numbers are off. Quick Summary:
- His ridership estimates are inflated.
- His engineering costs are wildly understated - thermal expansion of the tube alone kills his estimates, since any solution vastly increases the cost.
To be pendantic, that hasn’t been established.
The point of this thread was to discuss the engineering practicality of the hyperloop. Can it physically be built and work as described? If it needs changes to the design, are those changes possible with current technology without raising the cost to a degree that is unfeasible?
Economics is too hard to quantify because the very existence of the hyperloop would substantially change the number of people who travel between cities. Furthermore, if the technology were reliable enough, there might be many other places where it would make sense to build one of these.
Ultimately, social and economics problems are driven by underlying technological and engineering realities. People can adapt, physics can’t.
Sam Stone’s thermal expansion problem is the main “show stopping” problem discussed on this thread. However, proposed solutions (a third tube, expansion joints every 1 mile) are not “vast cost increases”, they still make the cost estimates for the system a tiny fraction of the estimates for high speed rail. (2-3 billion for another tube, we could pay 1 million per thermal expansion joint and that’s only another billion or 2, expensive but not unfeasible per say)
Well, to be doubly pedantic, the claimed ridership and engineering costs from the pdf, havent been established either.
I must have misunderstood your original post. You state above the point of the thread was to discuss engineering practicality, but your original post asks “How could this system fail?”
I understood this to encompass ALL areas where it could fail, not just engineering or technical areas. I apologize for misunderstanding that you only wanted technical reasons why it might fail.
With that in mind, I will re-iterate again, that I see no technical reason why we can’t build this. We can overcome all technical hurdles. We just don’t want to.
It won’t be built to the size, scale and numbers that are desired. Maybe, just maybe, there will be a pilot scale one built, but a system criss-crossing the country that is actually profitable…no, that won’t happen.
Why do you assume this? What need to travel between those cities is currently going unfilled? Also, his assumption is based on perfectly full capsules every 2 min, 24x7 in both directions. Traffic flow doesn’t work like that for ANY other transportation system and it’s foolish to assume that it would work that way for this one. This is all not to mention any downtime for the system.
The problem here is that your estimate for the 3rd tube is based on the assumptions for the first two are correct. If each tube is two or three times (or ten) times the cost, then suddenly the 3rd tube costs an extra 4-30B.
But let’s assume for a second that the darn thing could be built for 2-3B a tube (as currently designed). If we assume 1M per expansion joint with an expansion joint every 10th pylon that’s 2.5B PER TUBE. Now look at the pylon design. He only budgeted 100K per pylon, but that was without expansion joints and only two tubes, so beef that up. Let’s call it a 75% increase (too low still, but we’ll go with it). 25,000 pylons at an extra 50k a piece. There’s another 1.8B in cost.
You just doubled the build cost for a ZERO percent increase in ridership. So now we’re up to a 40 year payout (or a doubled ticket price - which will affect ridership - negatively).
Now note that all of the above assumes the original tube cost estimates were accurate, but let’s add a bit as a safety factor (after all, very few - if any - new tech engineering projects are built on budget). We’re at a cost of 5-6B per tube, and we need to add a cost overrun. Let’s say 20%. 6-7.2B per tube now. Where’s our payout now? 50 years? 60 years?
Ahhh, but we’ve also added maintenance, power, and equipment for a third tube (again with no increase in ridership). We also have to add in the maintenance costs of 7,500 expansion joints scattered over the length of the thing (not included in the original estimates). That means the original “net revenue” numbers are off. So we either have to increase price again or extend the payout some more.
See how quickly this blows up? And all of this is really only dealing with ONE of the engineering problems pointed out in this thread (thermal expansion). There are others (safety factors, capsule design, load/unload timing issues, etc).
I think you just described a subway.
There are always needs (or at least, desires) that are going unfilled. Maybe you’ve got a family in LA that likes to visit their grandma in San Francisco, but their travel budget is only enough for them to go once a year. If you put in a new travel option that costs half as much as what they’re using now, they’ll go visit her twice a year. Or, maybe it puts the cost below the threshold where a class of schoolchildren in one city can make a field trip to the other. Maybe someone’s willing to take a job in another city where he can come home to visit his family every weekend, but wouldn’t be willing to if he could only come home every other weekend.
You might want to read up on some of the trips people took when the budget airlines first got started, and offered loss-leader fares to capture market share. There were a lot of people doing travel that they wouldn’t have done were it more expensive. Were they wailing and gnashing their teeth about their inability to do it before? No. But they still found demand.
Yes, but not to 3x or 4x the previous demand. As noted above, the entire population of both cities would have to average 3+ trips per year to meet the numbers. And they’d have to 56 people at a time (28 in each direction) every 2 minutes 24x7. A class field trip alone would take about 1 full capsule in each direction. What happens to the passengers that otherwise would have taken that capsule? What happens on Saturday or Sunday when school is out? How do you handle the Giants playing the Dodgers where a whole mass of people want to go in one direction in a narrow time frame and then return all at the same time a few hours later?
Don’t forget this other thread that has also been discussing the technical feasibility, with Stranger piling heavy doubts on it. I agree with Sigene, though, that it was not at all apparent that we could only consider the technical end. Cost and usage are literally more important in determining feasibility. And without a working test model, it is pie-in-the-sky period.
Getting the two together is a deeply difficult problem. Go back in time and there’s no end of projects that are over- or under-estimated in cost or ridership. The future can’t be predicted. And the existence of the new project will change traffic patterns, often in totally unexpected ways. A best guess has to be sufficient or nothing would ever be built but while a product can be taken off the shelves if it doesn’t sell, transportation corridors persist. Building them requires satisfying high expectations, as it should. Limiting discussion to tech only is mere techy masturbation. If you want a real discussion, then everything has to be on the table.
I didn’t mean to imply otherwise. I was making the general point that usage would increase, not saying anything about how much it would increase.
And you can’t really separate the engineering and economic questions, since the engineering is going to determine how much it costs to build and run, and how much it costs to build and run is going to determine how much you can charge for a ticket in order to get a reasonable profit, and how much you can charge for a ticket (as well as things like speed, which is also an engineering question) is going to determine how many people ride it. If you could somehow engineer something with operating costs (including amortized R&D and construction costs) of $5 per person-trip that could get people from LA to SF in a half-hour, then you’d have no trouble at all with the economics… But you can’t do that.
If you read the pdf linked in the OP, Musk does mention expansion joints, as well as multiple pumping stations (for air evacuation), so he is not describing a single, solid tube 400 miles long. He also suggests that rush hour might involve capsules leaving at 30 second intervals (though, to me, it seems like some means to link four capsules and send them together would make more sense, preserving the 23 mile interval), I think he is putting forth the two minute interval as an average (there would just not be enough demand to shoot 30 capsules through between two and three in the morning).
Once again, while I complete agree that ridership estimates are ludicrous (we’re going to have full capsules at 3:14 AM and 3:16 AM and 3:18 AM every night of the year?), I think that thermal expansion is basically a solved engineering problem. Last I checked, the natural gas line to my house hasn’t ruptured, nor (in general) do the large pipelines that move gasoline and oil across the continent break every time the temperature changes. And, again, these lines all deal with pressures greater than one atmosphere.
Sure, they’ll need to account for thermal expansion, but – until a real pipeline engineer weighs in to the contrary – it seems to me to be something that’s been solved many many times before at reasonable cost.
One thing the PDF glosses over is the cost of acquisition of land. They say they’ll use the freeway for much of the route, to minimize land acquisition costs, but even so they way underestimate how much that costs.
My dad, who is now retired, used to work on rail transit projects, and he tells me the cost of acquiring land was always one of the biggest expenses, often bigger than construction or engineering. If the even the smallest length of Hyperloop is going through urban L.A. or Bay Area neighborhoods, that cost is huge.
Apart from ambient/environmental thermal expansion, what about heating caused by the magnetic fields? A steel tube is going to be ferro-magnetic, would the tube itself not cause significant energy losses in the system and suffer from some degree of degradation from the cycling electromagnets used to move the capsules? Is steel really the best material choice?
Keystone Phases I and II were built for 5.2B across 2,400 miles for an average cost per mile of 2.1M. That’s for 1 30-36 inch pipeline (113 inch circumference) averaging about 4ft off the ground. Musk is proposing building 2 “pipelines” of 88 inch diameter pipe (276 inch circumference) roughly 20-50 ft off the ground. So, double the number, more than double the size (2.44), and call it 6 times higher off the ground. And Musk is saying he can do all this for only around 7 times the cost. Hell, double the material for each pipeline and double the number of pipelines and you’re already at almost 5x without dealing with the height/weight on the pylons or solving any other engineering issues!
Further a few other key differences:
- The oil isn’t moving at 700+mph.
- The oil won’t die if it encounters turbulence.
- Oil pipelines are often built in zig-zags to deal with thermal expansion (a non-starter for HyperLoop).
- You don’t have to run a vacuum for an oil pipeline (safety issues with explosive recompression).
This like saying: “Well, we tunneled through a mountain, so the Chunnel should have been a piece of cake!”
Could you give an example? In my experience, there are more windows than rows of seats, and every person is either next to a window, or one seat away from an aisle.
I just looked at the propulsion (figure 10, pg. 18) as a thermodynamics example. Even assuming no losses, I can get no more than 46N of net thrust out of the system. This is quite a bit less than the 170N in the figure. Is there anybody willing to double check my calc’s?