Hacker News Reader

Comment by imaginenore | Hacker News Reader

I'm happy for them, but I'm yet to hear for anybody to address the major challenges that many people have raised. The major one being - the sudden pressurization in any point and the consequent shockwave traveling at the speed of sound, exploding the tube, killing everything in its way.

Thunderf00t did a good job summarizing the problems:

https://m.youtube.com/watch?v=RNFesa01llk

View on HN

I thought Thunderfoot's video has been pretty thoroughly debunked?

e.g. https://www.reddit.com/r/Futurology/comments/4udgd2/the_hype...

Also from Hyperloop One employees here:

https://www.reddit.com/r/IAmA/comments/4wck43/hi_were_mostly...

That doesn't seem like the main problem to me. The main problem is that the system as proposed by Musk, even taking all of the optimistic assumptions at face value, has a daily passenger capacity close to zero. The tube capacity is proposed to be less than 1000 passengers per hour. That doesn't even compare well with the capacity of a single lane of highway, and is orders of magnitude below what you get from a railroad.

Indeed . . . from the first announcement, I was intrigued by the concept and I truly hope it works out. HOWEVER, as proposed, it is no real competition with high speed rail. In fact, in my opinion, Musk missed the point of high speed rail entirely. It is less about getting from LA to SF quickly than it is about getting from Bakersfield to SF or from Fresno to LA. High speed rail links LA to SF as well as ALL POINTS IN BETWEEN. It creates a single economic corridor. Hyperloop on the other hand just links two points much the same way as air travel. This, coupled with the massive disparity in capacity that you pointed out, makes it ridiculous to compare Musk's hyperloop to high speed rail; it is more comparable to air travel.

Well, there could be a niche, upscale market for getting from SF to LA in 30 minutes, like people going for meetings and coming back on the same day. Other modes of transportation takes just too long for that kind of use case. I don't think this was ever thought of as a transportation option for people who commute to work every day because they can't afford to live in the center of those cities. Hyperloop, if it's ever built and works, will probably be the Concorde of ground transportation, in terms of market share and price. Although if we look at the outcome of Concorde, it may not make a lot of sense to follow that path.

Oh, I agree; there are use cases, but the original announcement billed it as a better alternative to high speed rail, and I just don't think it is in remotely the same category.

I think the real Musk's goal is to have companies start researching this technology and perfecting it, because it could be quite useful when colonizing Mars.

Ha! I think you're absolutely correct. An interesting twist.

> High speed rail links LA to SF as well as ALL POINTS IN BETWEEN.

Except when you add the 'all points in between' it stops being high speed rail. The current CA 'bullet train' as planned is projected to be only marginally faster than standard rail over the majority of it's run - 80-100MPH.

Accepting your statement at face value, that is still faster than driving and more convenient than the existing system which is, well, nothing. Standard rail, which does not currently exist on that route, would be even slower with the same stops. Los Angeles to Oakland on the Coast Starlight takes 11+ hours. [EDIT: Bakersfield to Oakland on the San Joaquins is about 6 hours. At 90 mph, high speed rail would cut that trip in half.]

Further, not every train has to stop at every station; there can be express trains.

Bakersfield to Oakland is a 4 hour drive by any reasonable measure, and it would be cheaper to boot.

The Cali bull*^% train isn't competing against rail - it's competing against cars and airplanes.

German ICE trains don't run at full speed except in certain segments. At full speed, they get over 300kph!

Also, you're forgetting about express trains. With sidings and in-cab signals, you can have high speed rail that stop at ever stations and others that pass intermediary stations for morning/evening commuters.

No, I'm not forgetting them. Part of the lower speeds projected are due to shared rail, and noise/safety concerns traveling through populated areas.

> The tube capacity is proposed to be less than 1000 passengers per hour. That doesn't even compare well with the capacity of a single lane of highway

It takes 3 to 5 airplanes departing per hour to get near that capacity.

Major airports handle departures at the pace of 30-40 per hour.

Airport capacity is not the point.

The point is that each plane costs around 300 million. A single modern high-speed train costs less than 100 million, and has over 5 times the capacity of a single passenger plane.

I don't think the shockwave would explode the tube. The in-rushing air would be equalizing the pressure. It would have a deleterious effect on pods, however. I would equip the pods with airbags that could inflate and act as emergency brakes, which would also compartmentalize the tube. The emergency protocol would be to bring all cars to a stop, then inflate the airbags. If the shockwave is impinging on the car, the airbag would inflate immediately. Then each compartmentalized tube segment would be rescued, with access through emergency access points. (Which could also contain the mechanism that takes care of the thermal expansion problem.)

I don't think the shockwave would explode the tube. The in-rushing air would be equalizing the pressure.

Yes, the more I've thought about this, the more I agree. Moreover, I don't think the hypothetical shockwave would last all that long. As it moves down the tube, it has to drag an increasingly long column of air behind it, all of which is likely turbulent and definitely causing friction with the inside of the tube. This would slow and weaken the shock front considerably to the point that at some distance down the tube, the pressure would rise not suddenly, but slowly and evenly.

Or the massive G forces on the passengers.

If you're going 650 Mph in an airplane and hit a bump (air pocket), the plane can bounce 5-10 feet. So the acceleration is spread over a long time / distance.

If you're going 650 Mph in a hyperloop and hit a bump, you have millimeters of room. So any minor bump is going to be a gigantic slam.

I've ridden TGVs in Europe. At 250 Km/h (~150 Mph), it's bumpy enough that I have a hard time doing anything other than stare out the window.

The hyperloop will be 4x faster than that. If it does work, I'd probably want a padded interior, and a face mask with attached vacuum aspirator for vomit...

> I've ridden TGVs in Europe. At 250 Km/h (~150 Mph), it's bumpy enough that I have a hard time doing anything other than stare out the window.

This is the total opposite of my experience, I've ridden the TGV many times in France and I'm always amazed how smooth the ride is, without looking outside I wouldn't even notice I'm moving. It's actually a weird feeling to look outside as you see the landscape moving so fast.

Same experience for me- I took the Eurostar from London to Paris (and back) last summer and couldn't tell you if we were moving or stopped, if I closed my eyes.

I used to live in France, and would sometimes commute from the south to Paris for the day.

Sure, there are large stretches where it is straight and smooth. But when it's bumpy, it's really bumpy. In a way that a slow train just can't do.

I expect it's "really bumpy" on the non-high-speed tracks, the train would be extremely rigid all around and thus not absorb bumps from low-quality/low-speed tracks making imperfections much more noticeable. My experience seems to have been mostly on proper tracks/at speed, the ride is as smooth as can be in a wheeled vehicle.

Which South? If it is via the Atlantic-side line, only the 200 km near Paris are a high-speed line, the rest is a classical line (on which the TGV rides).

I don't know what train you've ridden, but Spain's Ave goes 300 km/h and you don't even notice you're moving. And it definitely doesn't bump at all. You cannot notice constant speed but only speed changes (acceleration = force).

A better comparison might be the Transrapid Maglev in Shanghai. It goes 320km/h and it was incredibly smooth because it's levitating

> I've ridden TGVs in Europe. At 250 Km/h (~150 Mph), it's bumpy enough that I have a hard time doing anything other than stare out the window.

Uh? The only problem you may feel is the pressure in your ears when you are in a tunnel. But the ride feels much less bumpy than a regular train on a classical line.

> So the acceleration is spread over a long time / distance

That's not how acceleration works. Maybe you mean force? Hyperloop cars are heavy enough that a small air pocket wouldn't be able to produce enough force to accelerate it substantially.

Perhaps it is encouraging then to see this first go of it be for transporting cargo rather than people.

..like the transporters in Star Trek :)

Just as aircraft (and I'm sure other pressure vessels) are designed to resist explosive decompression, I am sure any real hyperloop system would be as well. That is not to say that it could never happen, but with proper engineering (and maintenance), we can make it exceedingly unlikely.

I'm not talking about the pod depressurization, which is indeed analogous to the plane, but is a separate issue, and is actually much worse than the plane (planes don't fly in near-vacuum, and can descend very quickly).

I'm talking about the tube pressurization, which is much much worse.

So am I. You can still prevent explosive decompression in much the same way. If you provide reinforcing strapping for example, you prevent a minor puncture from turning into a massive tear and explosive decompression [EDIT: perhaps, I should say re-compression]. I.E. if air is leaking in relatively slowly, you will never get the violent pressure wave you are talking about.

That doesn't solve larger tears, like a terrorist shooting an RPG at it, or exploding something underneath, or flying a small plane into it. It's a very large target to secure compared to a plane.

Fair enough. However, I think you are underestimating the size of explosion that it would take to seriously damage something like this. Engineers will definitely think about things like earthquakes and accidental (or intentional) collisions (such as a truck hitting a support pillar).

I also think you are overestimating the number of terrorists running around with RPGs.

EDIT: Of course, when all is said and done, the additional engineering requirements to make it safe may also make it economically unfeasible, in which case, your original point about the dangers would be essentially correct.

Unfortunately vacuum tubes are very vulnerable, they would have to be ridiculously overdesigned to take even a small explosion or a truck.

Terrorist with an RPG is just an example, it can be anything - a drone with C4, somebody throwing a long string of burning thermite, or just a gasoline canister set on fire.

All of these things can cause the same loss of life on any current mass transportation system.

This reminds me of the people who were very vocal about automated cars needing to be perfect before they could replace human drivers. They seem to have quieted down as the technology gets closer to actual deployment. I suppose that will also be the case here if the hyperloop becomes reality.

Not the same loss of life. A large rapture will propagate through the hole vacuum tube at the speed of sound, likely killing everyone, not just one pod.

Also it's a much easier target. Planes are only reachable to the terrorists during takeoffs/landings. And even then, it's not exactly easy to hit a plane moving at 300mph with an RPG/truck/drone.

The rest of your argument about automated cars is the classic example of straw man.

You see, I didn't actually argue about automated cars, just noticed the similarity in the arguments.

And to address your original point in a horribly sarcastic way, if only someone had invented bulkhead technology...

Our current transportation system also doesn't solve terrorists planting bombs on rail lines or shooting RPGs at train cars or crashing planes into highway overpasses.

Not the same scale of destruction, not even close.

You might derail one train, if lucky, not every train on the track. We have defect detectors:

https://en.wikipedia.org/wiki/Defect_detector

Also trains, especially in the US, are much slower, and derailing one will probably kill a few passengers, but most will be fine. It's much more damaging to target bridges, which terrorists did:

https://en.wikipedia.org/wiki/Rafiganj_train_wreck (130 dead)

Crashing a plane into a highway overpass will do even less damage, because cars can stop very quickly when they see a problem.

Would building it underground largely solve these risks? We already build for natural disasters in mind- seems earthquakes would be the main issue in California, which structures claim to be "proofed" from.

The pressurization wouldn't be "explosive," there wouldn't be a pressure wave because a gradient of pressure would have to form between the atmosphere and the tube. Here's a playlist of videos rebutting this argument and many others:

https://www.youtube.com/playlist?list=PLSPi1JFx4_-Gz0Fm0qq2K...

You are correct in that it can be addressed, but I believe the issues people raise are that properly addressing everything makes the project to be (potentially) extraordinarily expensive.

Rebuttal playlist addressing all of his videos: https://www.youtube.com/playlist?list=PLSPi1JFx4_-Gz0Fm0qq2K...

Reply on news.ycombinator.com