Hyperloop

[rabar10]

For balance, here's an article that is critical of the Hyperloop proposal, not for technical reasons per se, but because of an unfair comparison to the CA high-speed rail project: Hyperloop proposal: Bad joke or attempt to sabotage California HSR project?

While I don't think the article title is accurate (i.e. this is some nefarious political plot) it brings up some valid points:
- Hyperloop doesn't reach downtown LA and so avoids the most expensive land rights issues
- B/c it would take more time to get from downtown in SF/LA to the proposed station locations, the overall time savings is not as attractive as proposed on the surface
- Downplaying the NIMBY-ism and "what's in it for me" nature of building any large transportation structure in CA

 

[Johan]

Any thoughts on what a demonstration project would look like?

My two cents:

- Not required to carry passangers, nor solar panels
- Should probably have a tube of full diamater (i.e. 7'4")
- Should reach the max velocity (760mph), and maybe coast for a mile or so
- Should maybe have curves of (close to) the minimum bending capacity, to test the banking in the curves and measure actual g-forces (centripetal)

The length of such a thing would depend greatly on the applicable acceleration/deceleration. Assuming 1g, you would need approx. 8.5 miles. At 2g you need approx. 5 miles. At 3g you need approx. 3.5 miles.

Generally assuming prototype has triple cost of full scale (this may be underestimating things) and using the Alpha design figures, the budget would be:

- 2x vehicle: 3 * 2 * MUSD1.35 = MUSD 8
- 5 miles of tube = 3 * 5 * MUSD 650/350 = MUSD 28
- pylons for 5 miles of tube = 3 * 5 * MUSD 2550/350 = MUSD 109
- propulsion (1 unit, compared to 10 in full scale): MUSD 140/10 = MUSD 14
- 5 miles of solar panels and batteries (3 * 5* MUSD 210/350) = MUSD 9
- Station and vacuum pumps (MUSD 260 for full scale): MUSD 10?

Total: MUSD 178

I think it would make no sense to put the prototype on pylons. Probably it could be put on simple concrete foundations. That could bring the "pylons" cost down by 90% or something. In that case we get a figure south of MUSD 100.

Excellent break-down. In reality not a very big cost if that would be what it takes to create proof of concept. If one were to build such a prototype, I think it should be built with capsules that do carry passengers and are not far from the proposed final specifications when it comes to design and comfort, since that will be an important selling point in order to gain acceptance and financing for a full scale project.

 

[DonPedro]

Excellent break-down. In reality not a very big cost if that would be what it takes to create proof of concept. If one were to build such a prototype, I think it should be built with capsules that do carry passengers and are not far from the proposed final specifications when it comes to design and comfort, since that will be an important selling point in order to gain acceptance and financing for a full scale project.

Good point. You would do a gazillion test runs first, and then add passengers once it was clear that all the life support and other security features work as planned. You would have to add length of track, as I doubt you would want to put the test passengers through 20 seconds of 2g acceleration, followed by 20 seconds of 2g braking. With a 1g limit, you get an 8.5 mile track which costs ~MUSD 33 extra with the assumptions above (including 90% saving off pylon costs by putting on ground).

 

[IceWendigo]

" B/c it would take more time to get from downtown in SF/LA to the proposed station locations, the overall time savings is not as attractive as proposed on the surface"

I dont know about SF/LA in particular, but in many large cities, if your starting point is outside of the downtown area, going downtown where a train station is can also be time consuming.
If you startoff outside of downtown in the morning rush hour, getting there can be more of a headache than going to a location in the periphery.

What would be great is a Miniloop elevated urban/local system (or something like SkyTran) going from the downtown and a few points to the Hyperloop station

 

[Norbert]

Perhaps, to save money, the first prototype would be "subscale", for example 1:5, and be designed to run in a tight curve, so that one can build a circle. This would allow to test full half-hour runs.

 

[JRP3]

Agree, subscale mini loop for the initial testing, maybe even toy sized. Cheaper, faster to build, and all things can be monitored, including forces and temperatures that passengers might be subject to, and then scaled up. Could be the next cool toy instead of model trains :biggrin:

 

[DonPedro]

Makes sense. You can learn a lot from that at a cost that is one or two orders of magnitudes less. You would probably still need to build the 1:1 scale prototype, but it would be cheaper and hit the nail more squarely on the head.

 

[kevin99]

Agree, subscale mini loop for the initial testing, maybe even toy sized. Cheaper, faster to build, and all things can be monitored, including forces and temperatures that passengers might be subject to, and then scaled up. Could be the next cool toy instead of model trains :biggrin:

If it is running at 700+ miles, you won't be able to where the capsule is at... It is an invisible toy!

 

[IceWendigo]

There's something Id find psychologically unpleasant. I remember a story in Pearl Harbor about a ship having rolled as it sunk, and people were trapped by the hull.

In the very unlikely scenario where a part/piece of pod is broken off and causes damage on the next pod that disables it and causes a leak... If you are trapped mid way, you are in hostile deadly environment trapped in steel, and unlike a plane that could dive down to normal pressure, though normal environment is a few feet away you are trapped, I dont know if the pod door even open, and even if you could exit the tube environment is deadly, and its not like there'd be an exit 20 feet away, it might be miles away.
If something happens and a pod is immobilized many miles down the tube, what is the emergency response procedure?

 

[gregincal]

For balance, here's an article that is critical of the Hyperloop proposal, not for technical reasons per se, but because of an unfair comparison to the CA high-speed rail project: Hyperloop proposal: Bad joke or attempt to sabotage California HSR project?

While I don't think the article title is accurate (i.e. this is some nefarious political plot) it brings up some valid points:
- Hyperloop doesn't reach downtown LA and so avoids the most expensive land rights issues
- B/c it would take more time to get from downtown in SF/LA to the proposed station locations, the overall time savings is not as attractive as proposed on the surface
- Downplaying the NIMBY-ism and "what's in it for me" nature of building any large transportation structure in CA

Yeah, I do think the elevated tubes would get huge push back in urban areas. I5 is mostly rural, but as they point out the Central Valley section isn't what makes the current high speed rail expensive or slow.

 

[ElSupreme]

Hey Elon you also stole this idea from two of the greatest minds of our time, Rage Kage (aka: KG, or Kyle Gass) and Jack Black.

This really has some foul language. But I can't help myself but to think of this song when reading about the hyperloop. I have advanced to the correct part, it takes about 10 seconds for the payoff.

 

[stevezzzz]

There's something Id find psychologically unpleasant. I remember a story in Pearl Harbor about a ship having rolled as it sunk, and people were trapped by the hull.

In the very unlikely scenario where a part/piece of pod is broken off and causes damage on the next pod that disables it and causes a leak... If you are trapped mid way, you are in hostile deadly environment trapped in steel, and unlike a plane that could dive down to normal pressure, though normal environment is a few feet away you are trapped, I dont know if the pod door even open, and even if you could exit the tube environment is deadly, and its not like there'd be an exit 20 feet away, it might be miles away.
If something happens and a pod is immobilized many miles down the tube, what is the emergency response procedure?

The ability to rapidly and non-destructively re-pressurize the tube is a requirement (the analog of diving to lower altitudes in a commercial jetliner). At that point, since the main cabin doors probably can't be opened in the confines of the tube and there's no room to exit between the capsule and sidewall, there would have to be an emergency hatch in the roof of the cabin. Passengers would be told by radio when to open the hatch, clamber out and start walking. I imagine the re-pressurization vents would have to be spaced reasonably close to each other, and could be designed to double as emergency exits, with fabric escape tubes like you may have seen in other applications, to get people safely to the ground.

 

[Rodolfo Paiz]

A fellow I know used to have a private plane that was pressurized. He had an intermittent leak problem somewhere and had a heck of a time getting it resolved. I do believe you have to avoid leaks to a fairly significant degree...

Finding and resolving individual leaks can be time-consuming and annoying. That being said, leaks are a very small and infrequent issue with aircraft pressurization systems: it's neither difficult nor usually threatening. I'm involved with the operation and management of two pressurized general-aviation airplanes, and have been for years... it's just not a problem. Takes work and time, but does not introduce any significant risk.

While the risks of decompression are very real, providing a safe environment within the Hyperloop shouldn't be too hard. The pod's passenger and baggage compartments are a sealed bottle, essentially: you load the people, you close it up, you're done. If you have a minor leak, you use some of the reserve air to compensate and maintain pressure. All you have to do is define the loss of air (easily measured in air mass consumed from the reserve) at which a minor leak flags the pod to be taken out of service for maintenance after this particular trip. That's not hard at all, and it would cover some large (way over 99%) fraction of cases.

You'd only need the masks if the leak was significant enough that the reserve air supply couldn't keep up and the pod ended up losing 31% of its air pressure (down to 10psi/70kPa, roughly the equivalent of 10,000 feet of altitude) within the length of a single trip. That's a hell of a leak, and should hardly ever happen; especially since you can design into the pod the ability to resupply itself with large amounts of air from the reserve to reduce the odds. Even here in the "minor leak" realm, the probability that any given pod will need to drop the masks is minuscule: if it happens four times a year on the SF-LA loop, which would still be (IMHO) unrealistically high for a well-designed pod traveling inside a protected tube, you're talking about a 0.0000025% chance (1 in nearly 400,000).

In that rare case where you did have to drop the masks at 10psi, you still have a big margin between that and 2psi/14kPa where things get life-threatening. The system operator would set a hard limit at which emergency repressurization of the tube is undertaken, and I'd say (for the purposes of this discussion) it would be something like 6psi/41kPa to enable the tube to reach a reasonable pressure of at least 3psi/21kPa before the pod ever gets to a life-threatening situation. The key problem to solve here, which is not egregiously difficult, is to determine how quickly the tube can be brought up from its standard operating pressure to 3psi/21kPa without incurring unacceptable effects of shockwaves, friction heating, and so on, and then set your safety parameters that way.

Given a good design effort, I just don't see loss of pressure as being a highly-difficult problem. There are other things, like the whole expansion-joint issue, which are thornier.

- - - Updated - - -

In the very unlikely scenario where a part/piece of pod is broken off and causes damage on the next pod that disables it and causes a leak... If you are trapped mid way, you are in hostile deadly environment trapped in steel, and unlike a plane that could dive down to normal pressure, though normal environment is a few feet away you are trapped, I dont know if the pod door even open, and even if you could exit the tube environment is deadly, and its not like there'd be an exit 20 feet away, it might be miles away.
If something happens and a pod is immobilized many miles down the tube, what is the emergency response procedure?

Remember the pods are also equipped with wheels to drive themselves forward, and the tube can be repressurized so it's no longer a lethal environment, just one where the pods can't go full speed. Foreign-object damage such as you mentioned is a major concern during design, to make such a "piece breaks off a pod" scenario nearly impossible. In the strange case that it does happen and does leave a pod stuck, Steve's right and the nearest escape hatch is probably a few pylons away.

 

[DonPedro]

Regarding risk of leaks, it is good to keep in mind that the differential pressure is <1 atmosphere, which is comparable with a submarine at 30 feet water depth. (Yes, a submarine has a thicker hull, is more expensive and is not moving through near-vacuum at 700mph - just wanted to point out that the pressure to be contained is not very big).

 

[IceWendigo]

The idea of emergency re-pressurization and exits are reassuring.

If prototypes are proven and perfected, maybe instead of 'a priori' LA-SF project(as great as it might be for CA), it might be interesting, (potentially with a bit of corporate and crowd-source financing) to consider turning the tables around and have several regions around the world, bid to have the first project. Since Europe, Japan and China have already taken the lead on High-speed trains, maybe another country would gladly want to collaborate in the implementation of the Hyperloop.

 

[jeff_adams]

The idea of emergency re-pressurization and exits are reassuring.

If prototypes are proven and perfected, maybe instead of 'a priori' LA-SF project(as great as it might be for CA), it might be interesting, (potentially with a bit of corporate and crowd-source financing) to consider turning the tables around and have several regions around the world, bid to have the first project. Since Europe, Japan and China have already taken the lead on High-speed trains, maybe another country would gladly want to collaborate in the implementation of the Hyperloop.

If that were to happen, I'd wager Norway would be the first. The populous might be infatuated with Tesla by then and they would be most likely to try something like this.

 

[Rodolfo Paiz]

I think the SF-LA loop was designed for the alpha doc to provide a direct comparison to California's high-speed rail project. Nothing says that this should be first or that they expect it to be first. Anyone who wants to attack this problem is free to do so without any bidding: the whole idea is open-source.

 

[RDoc]

For balance, here's an article that is critical of the Hyperloop proposal, not for technical reasons per se, but because of an unfair comparison to the CA high-speed rail project: Hyperloop proposal: Bad joke or attempt to sabotage California HSR project?

While I don't think the article title is accurate (i.e. this is some nefarious political plot) it brings up some valid points:
- Hyperloop doesn't reach downtown LA and so avoids the most expensive land rights issues
- B/c it would take more time to get from downtown in SF/LA to the proposed station locations, the overall time savings is not as attractive as proposed on the surface
- Downplaying the NIMBY-ism and "what's in it for me" nature of building any large transportation structure in CA

I too thought the criticism that there are no plans or cost estimates for how to actually get into the cities and build stations in them is very fair. This should be addressed, both how it would be done and what the cost would be. One of the main advantages trains have over planes is they stop in the center of cities and without that, Hyperloop loses a lot of its attraction.

The criticism that politics and NIMBY isn't fully addressed is also a serious issue. Without any intermediate stations, why would any of the local governments approve the plan? Running it down I5 might help a lot, but there still could be a huge political problem if there weren't a way to get the local governments on board. Perhaps grants to local communities during construction and an added fee on the tickets for local viewscape disruption would help. I don't have any idea how much this would cost, but it's certainly not zero.

There's a similar criticism at http://greatergreaterwashington.org/post/19848/musks-hyperloop-math-doesnt-add-up/. That article's point is mainly that the maximum throughput of the Hyperloop is much less than the proposed train system, and that the Hyperloop's capacity is false because the car spacing has to be greater for safety. I have no comment on the train ridership projections except that they are quite controversial. However, the assertion that the spacing between cars would have to be increased for safety is at best disingenuous IMHO. The assumption is that the spacing can't be less than that required to stop a car, so at .5 g the spacing would have to be increased to 80 seconds, thereby decreasing the hourly capacity. Assuming that the emergency stopping rate is the same as normal stopping seems ridiculous to me since in an emergency, passenger safety would trump passenger comfort. There are many street legal cars that can safely stop at over 1g and, with a solid metal tube surrounding the Hyperloop cars to push against, an emergency braking system that stops at the required 1.2 g seems perfectly save and reasonable.

 

[DonPedro]

If that were to happen, I'd wager Norway would be the first. The populous might be infatuated with Tesla by then and they would be most likely to try something like this.

So far, I think less than 0.1% of the population is infatuated with Tesla, so there is a way to go there. But I am going to do the numbers and see if this is worth advocating. One the positive side, we don't have to look at seismic loads, and we have a government that just might fund something like this. On the flip side, we have much smaller cities to connect (the two biggest ones having 900k and 400k inhabitants). We also have a lot of mountains and valleys between the cities, making it hard to get straight lines without a huge amount of tunneling. I'm curious how it could work out.

Oh, and we would go without the solar panels.

 

[gym7rjm]

The idea of emergency re-pressurization and exits are reassuring.

If prototypes are proven and perfected, maybe instead of 'a priori' LA-SF project(as great as it might be for CA), it might be interesting, (potentially with a bit of corporate and crowd-source financing) to consider turning the tables around and have several regions around the world, bid to have the first project. Since Europe, Japan and China have already taken the lead on High-speed trains, maybe another country would gladly want to collaborate in the implementation of the Hyperloop.

Definitely agree, LA-SF path seems like there are too many politics surrounding the project for it to easily get off the ground. I would love to see something like the Google Fiber project where cities pairs would bid for it. I think there could be a better option than LA-SF where land rights don't matter as much and where it could be mostly ground based. Maybe for example Dallas - Houston? Its also a much shorter route so it would require less length in track and fewer pods. I haven't done the math, but seems like it could cost half of SF-LA. It would also provide an excellent proof of concept for future city pairs to go off of.

And speaking of Google and the Fiber model... they have loads of cash and seem to go after exciting projects, this seems like a perfect match for them. They could even tie it in to their advertising business model. Imagine rolling onto the hyperloop and having 5 minutes of ads streamed to your 17in touchscreen display? or pay a little extra for an ad-free ride. And this would cost Google less than the Motorola acquisition!