Hacker News new | comments | ask | show | jobs | submit login
I Think the Hyperloop is a Maglev Tube Train (dylanized.com)
11 points by dylanhassinger on Sept 17, 2012 | hide | past | web | favorite | 19 comments

Some of the points seem a bit silly.

+ For instance saying it "has no tracks" but then ... it's in a tube. This seems to be nitpicking, as the tube is effectively a track. The question of whether something uses a track or not is mainly of interest because things that don't use tracks have more flexibility, and can go off-route. Something that requires a fixed guideway (like a tube) is for all intents and purposes, on a track.

+ Most importantly, the tubes would considerably safer than traditional high speed rail, since debris or other obstacles couldn’t get in and block the vehicles.

Er, is "debris and obstacles" an actual safety problem on existing HSR lines? Oh wait, Shinkansen: perfect safety record over 50+ years of very intense operation.

It's pretty clear that grade separation, and some effort to avoid trespassers, are important, but these things are of course standard practice. Putting everything in a tunnel would appear to at best offer a minimal safety improvement, at great additional cost.

+ Similarly, the vague hand-wavy discussion saying that the tubes would be "cheap and easy to construct" seem a bit specious. Either precision is required or not (and at 500+mph, one had better get this right!), and the form of a tube doesn't seem to make much difference compared to other maglev technologies, e.g. the "U-channel" being used by the JR maglev (one of the notable attributes of the JR tech is that it has a very large gap size, and so is less susceptible to precision issues, as compared to e.g. the transrapid tech).

So if you take existing JR maglev tech, and ... add a roof, is it actually any different than the vague "tube" notion this article talks about?

Maybe Musk does have some truly transformative idea, but certainly nothing in this article is that.

[I think Musk's coyness about this is pretty annoying... if he really has a great idea, and intends to make it public, why not just say it instead of making flirty glances and beating around the bush?]

> I think Musk's coyness about this is pretty annoying... if he really has a great idea, and intends to make it public, why not just say it instead of making flirty glances and beating around the bush?

I agree, and it reminds me of nothing so much as the protracted and anticlimactic unveiling of the Segway.

A tech-world-famous entrepreneur coyly keeping mum about some radical new technology that is supposed to CHANGE THE FREAKIN' WORLD, MAN--and then it turns out to be a revolution in keeping airport security guards from having to walk so much.

I'd love to see it be everything Musk promises (just like I'd love to order a personal cold fusion generator from Amazon), but such lofty claims need substantial evidence to have any credibility.

To be fair, he's mentioned it as an idea and I believe in the interview, he said he wanted to check with his engineers to make sure it could be possible.

He's not hyping up the next product as much as he's musing about the notes & ideas written down in his journal when people ask him bluntly "what else are you working on?"

And to be fair, Musk's track record on creating things that are going to CHANGE THE FREAKIN' WORLD, MAN is pretty dang good (paypal, tesla, spacex as if I have to even name them).

The feeling I get, and it's just a feeling, is that it's an idea where the hype is sort of getting away from him.

I wonder if he even took it all that seriously at first, he may not have realized he's made the transition to Man Whose Words Are Given Great Attention.

He seems like he's starting to realize that, but only now...

Musk: "oh crap!" ><


> I think Musk's coyness about this is pretty annoying... if he really has a great idea, and intends to make it public, why not just say it instead of making flirty glances and beating around the bush?

I would also say that a lot of great ideas when not executed carefully, frankly suck.

With Musk's precision about his products (checking every car himself), I doubt he's going to just say "here's my idea, have at it" without a built in way to implement it, make it happen while maintaining some semblance of control.

A closed tube could maybe have negligible air friction, if you can keep it moving.

An evacuated tube could have negligible air-friction—but an evacuated tube is very different than what's being discussed as it meets none of the apparent criteria (low-cost, low-maintenance, easily constructed, precision-insensitive).

Also, Musk explicitly said it's not that ("But Elon has said there’s no vaccuum tubes involved").

Has he ruled out running the trains in a lighter gas like hydrogen or helium?

Interesting, but then you have to worry about keeping oxygen in the train, and not having leaks?

Nah I am saying move the air at the same speed as the trains.

You still get friction then, just between the air and the tube walls.

The bad effects of air-friction would seem to still be present, and maybe much greater, because you'd have to move a huge mass of air and the friction would be over a much greater surface area (isolating the moving air to within a small portion of the tube does not look simple to me).

I tried a calculation. It appears there is an optimal gas velocity in between 0 and V_train. It depends on a lot of factors that I didn't spend much time trying to come up with realistic values. If I weight things heavily in my favor, you can save over 50% power requirements by moving the air at around ~120m/s. If I weight things in your favor there is a pretty paltry power savings at around ~10m/s. If you care, here was my method:

f=0.006; %wall friction factor inside tube, also skin friction outside a train rho=1.2; %air density [kg/m^3] R=1.5; %tube radius [m] Vg=0:200; %velocity of tube gas [m/s] Vt=250; %velocity of train [m/s] L=500e3; %length of tube [m] Cd=0.8; %drag coefficient, form drag trainfill=0.1; %percent of tube filled by train airfill=1-trainfill; %percent of tube filled by air ncars=1000; %number of cars in tube Pg=pi/4fVg.^3LairfillR; %power required to move gas through tube Pt_fm=ncars0.5rhopiR^2Cd(Vt-Vg).^3; %power required to overcome form drag on each car Pt_sk=pi/4f(Vt-Vg).^3LtrainfillR; %power required to overcome skin friction drag Pg_annulus=0; %since the mass of air is so small between the train and tube wall, assume 0. Pt=Pg+Pt_fm+Pt_sk; %total power requirements

Also, this article mentions that it is a tube of some sort: http://www.businessweek.com/articles/2012-09-13/elon-musk-th...

Constant starting and stopping would be a performance and efficiency issue. A better way would be to have local trains that ran on parallel tracks and sync'd speed with the main train and passengers and cargo would be xfered to the local train - alternately, traffic could be segregated into seperate cars by destination and then cars would be spliced back and forth between the local and main train.

It'd have to be a vented tube, or else you'd be spending most of your power shoving air through the tubes.

The per-kilometre cost is going to be astronomical.

I don't think one of the problems of maglev is stopping distance: that's mostly constrained by how many G's you want to subject your passengers to.

It might be possible to build a tube system with a blowback mechanism so that the displaced air can be temporarily pushed aside without wasting too much energy.

Most long train tunnels have a secondary tunnel that acts as a buffer for this sort of thing.

First, let me say I think that estimates of cost savings over the European/Japan figures are ridiculous for anything at the ground level that moves a lot faster than state-of-the-art maglev or HSR.

A closed tunnel doesn't buy you much.

Mostly-evacuated tube, or at the very least some type of sealed hydraulic (which is problematic) is the obvious plausible idea to achieve supersonic speeds. The problem is structural engineering of a very large negative-pressure-sealed structure in which a fault anywhere could destroy the whole thing. It would need complex airlocks, which would make each station enormously expensive as well.

I would note, however, one of my ideas that may have some synchronicity: that the speed of sound in a gas rises rapidly as the molecule gets less massive, and the drag force drops as the gas becomes less dense. Simultaneously, if hydrogen cars are going to get off the ground they need a distribution pipeline system. A train that traveled in a tube of hydrogen could effectively bypass the 'air' speed of sound and move at at least mach 3 before reaching the H2-transonic range. Sealing STP hydrogen in a container can be done an order of magnitude easier than sealing vacuum or highly pressurized gas.

The "tracks", of course, would be some form of maglev. Ideally, an 'Inductrack' design, which have not been seen in use yet but offer substantial advantages. On the other hand, given the progess on high-temperature superconductor manufacturing, perhaps those advantages are shrinking.

On Googling, it would appear that I'm not the only one to have this notion: http://www.supersonictubevehicle.com/docs/STV%20--%20Denizli...

I don't understand attempting to use a propeller-based primer mover, however. A linear induction motor with variable coil spacing, on a multiple unit train, should function just fine for this type of use.

> I started thinking it was some type of high-altitude blimp network, surfing low-earth orbit and dragging capsules with commuters inside

I'd like to hear more about that idea.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact