Hacker News Reader

Comment by raphaelj | Hacker News Reader

The problem with hyperloop is that is does not solve the main problem of bullet trains: infrastructure costs. Nobody complains about the TGV/Shinkansen/ICE being slow.

Building a new train line cost millions of €/$ per km/mi because you need to expropriate properties, build tunnels and bridges, fine tune the tracks (any small distortion in the tracks will cause significant annoyance for the travelers). Hyperloop will have all these issues, but worse, as the speed is higher, and as the hyperloop trains will not be able to run on existing tracks.

View on HN

German here. ICE sucks majorly.

ICE is engineered way over the limitations of the underlying technology. They break down easily and are constantly late.

I'd take a hyperloop over an ICE any time.

We had good tech, that that didn't get held back by the physical constraints of rail travel: Transrapid

A maglev train developed in the 70s which operated flawlessly for 40 years.

Sadly there was no political willpower for it's realisation due to a very strong car lobby.

I see the hyperloop as a modern spin on the Transrapid, and I'd pour all of our subsidies for Deutsche Bahn into it in a heartbeat.

German here. ICE rocks. Until like 2 years ago, I used to travel long distance nearly every week, often all the way across germany.

The problem is not the ICE or high speed rail connectivity, the problems arise at the edges, when you need to switch over and travel the last few miles. For example, I used to travel from Berlin to Hamburg for on-site work. The leg from Berlin to Hamburg was faster than the connection from Hamburg Main to the client site.

Certainly, there’s delays, but I’ve done the same trip by car and it constantly took longer than by train, I had very few trips where I didn’t end up caught in a traffic jam somewhere on the autobahn. There’s a bit of a perception problem: nobody notices if you’re late by five minutes if traveling by car because no one has such an exact expectation for car travel.

Delays in train travel also have a problem that is exacerbated by the timetables of the DB: In an attempt to make travel times seem shorter, there’s often insufficient buffer at places where you need to change, so that a few minutes of delay will result in an hour of layover. Combine that with lack of maintenance on the rails and you get unfavorable results.

None of the problems would be solved by hyper loop or any other technology. It’s a result of misaligned incentives.

Still, with all the things wrong in planning, maintenance and political incentives around it, high speed rail still is the best way to travel medium and long distance.

You neglect the timing advantages gained from smaller independent crafts.

Hyperloop has the ability to essentially do driverless packet routing.

You can't do that with high speed rail.

How would any of this be impossible on rail? None of that depends on “pods need to move in a tube”. And if it’s such a major benefit, why hasn’t any of this been done on light rail? There’s no driverless light rail anywhere in germany - and the question you need to answer is “why is that?” You’ll quickly figure out that some of the reasons are technical and some are political, but “must be in a tube and maglev” is not one of them.

Hyperloop appeals to us technical folks because it’s cool. But it suffers from many of the same problems that occur when technical folks propose solution, we tend to propose technical solutions to non-technical problems. And we usually fail because we ignore the non-technical side.

For safety concerns, you'd probably have to divide the route into air-locked sections with a pod per block. This gives you about one packet every 25 miles or so – and rather slow packet routing (more like RFC 1149 – https://tools.ietf.org/html/rfc1149 )

At 1200km/h thats one every 2 minutes per 40km (25miles).

So you get 30 pods per hour.

With 28 passengers each.

Which is 840 passengers per hour.

Compared with the different ICE 4 models, even the Largest ICE transports less passengers per hour.

499 (K1n) 724 (K3s, 10-car) 830 (K3, 12-car)

Sure, you might say that those are only the planned stats for the hyperloop. But even halfed it's still competitive due to speed alone.

Furthermore ICE trains are half empty most of the time, even when it's packed. Why? Because DB insists on having 18th century style 1st and 2nd class tickets. And most people can't afford that.

However hyperloop has no such distinction, because all the pods can be made to "first class standards" due to the direct point A to point B routing nature. Alternstively, the richt people have to wait until their pod is at capacty, which I'd be delighted to see.

So in reality ICE capacity is probably half of what hyperloop could provide.

Edit: I'm using the assumption of 1 ICE per hour, because that's how the entire german train system is sheduled. And we're comparing Hyperloop to the current system.

Seriously, taking per train numbers as hourly limits? And why would hyperloop magically be cheaper?

Even if it was the same price, it's better in every other metric.

It's cheaper because you can prefab parts that are otherwise not prefabbable due to the high weight of trains, and that require less maintenance due to not being exposed to the elements.

The entire german train system is build on crushed stone. With mechanically tensioned electrical lines that are in constant abrasive contact with the train. The main cost here is not the crushed stone, but the workforce required to set everything up.

Having the thing run as a highline with integrated contactless propulsion made from robotically mass produced segments, would be a lot cheaper, and a lot easier to maintain.

In theory you could even build the tune in a factory and just push it out, with each pylon containing a small pusher system. That way you can just produce it at the station and only need a groundgrew for the pylons and for alignment.

If you cover it with PV it'll even be energy net positive.

The price we save on CO2 and climate change alone would be worth it.

50% of ICE electrcity is still coal and will be for the coming decades. (Datteln 4)

Now there could be 10 ICEs passing a given track an hour, with various endpoints…

Yeah there could but that's not how ICE are sheduled most of the time. But there could also be 4 tubes stacked, or your security distance reduced.

Your whole capacity argument is based in the "acceptable amount of lives lost".

If you're ok with the same amount of lives lost in a catastrophic hyperloop accident, as with a catastrophic ICE accident and assuming that everyone perishes in case of a hyperloop accident (unrealistic but let's go with it), then that would be 101 lives. Which gives us 3.6 times as much capacity for the hyperloop, or 3030 passengers per hour per tube.

Say you stack 4 tubes, that's 12120 passengers per hour between two points compared to 8400 passengers for your ICE case.

And that still completely ignores the multi rout load balancing, last mile convenience, small village connection and no waiting time advantages gained from packet routing.

I'd take a hyperloop over an ICE any time.

Since a “hyperloop” doesn’t exist and you have no basis for understanding what the tradeoffs are, this is not very convincing.

The point remains - high-speed rail is a known and deliverable system. Good implementations exist in several countries. In contrast, a new technology comes with associated engineering and economic challenges, and it’s unclear how those can solve the existing problems that high-sped rail deals with.

Ok let me rephrase, and actually make my point stronger.

I'd take a shot at fully funding hyperloop with all the ressources that our current HSR system has, over keeping our ICEs any day.

If it was up to me we'd fund Transrapid. It's also build on pylons, does 500kmh, and is proven tech that has been running for 40 years. So we get the best of both.

But I don't see a future for Transrapid, because it's not cool enough for political backing.

Hyperloop is, so I'm doing a strategic vote here.

> It's also build on pylons

High-speed rail can also be built on pylons.[1]

The hyperloop white paper argued that being able to build on pylons would be an advantage of hyperloop over conventional rail:

> The key advantages of a tube vs. a railway track are that it can be built above the ground on pylons and it can be built in prefabricated sections that are dropped in place and joined with an orbital seam welder. By building it on pylons, you can almost entirely avoid the need to buy land by following alongside the mostly very straight California Interstate 5 highway, with only minor deviations when the highway makes a sharp turn.

The author seemed completely unaware that conventional rail can also use pylons, and can (and does) use prefabricated elements.

The whole white paper seemed terribly uninformed about the realities of high-speed rail.

1. http://img2.chinadaily.com.cn/images/201801/02/5a4b62cea3100...

The Transrapid was simply way too expensive. The planned route between Hamburg and Berlin, which is not even 300km, was shelved because the estimated costs exploded. The DB (German Railway) gets 2.5 billion EUR funding per year for its train network, which is 38.500km, so it's really not that much. The Transrapid wouldn't even have been significantly faster, since there were three stops planned between Hamburg and Berlin. And both the Transrapid and the ICE were/are built by Siemens, so there's not reason to think they would have been more reliable.

To add to that, one of the major problems was that due to the political power structure (state vs federal government etc.), a no-stop trip on the direct route wasn’t viable political. So the DB basically opted for refurbishing the existing track, because that did not need political approval on any level - the track existed and no new permits were required.

All of this would be issues that hyperloop would have to face as well. None of these are in any way related to the underlying technology.

In addition, there’s extremely few routes in germany where 300km/h vs. 500km/h top speed makes a significant difference. The latest generation of ICE trains trades a lower top speed for faster acceleration. We’re a small country with comparatively short distances between population centers.

Transrapid tech has no moving parts apart from switches and trains.

It's a lot easier to get that reliable.

Given that one of the major problems with the ICE is that often times the air conditioning doesn't work, I seriously doubt that. The real problem here is that you're dealing with a monopoly. Siemens simply has no reason to care, so they don't. It doesn't matter if they build an ICE or a Transrapid, it's not like the DB could switch suppliers. The problem in Germany is not to travel between major cities. The real issue is if you have to reach a city which is not big enough to be connected to the ICE network, since missing your connection often times means waiting for an hour.

Do TGV and Shinkansen have the problems you say ICE has?

Do they run in a literal vacuum?

> I'd pour all of our subsidies for Deutsche Bahn into it in a heartbeat.

congratulations, you just got a few high-speed links between half a dozen cities at the cost of rail infrastructure for the rest of the country (a large part of the subsidies goes into boring regional services). The answer to "our rail infrastructure should be better maintained" is not "dump all the money in shiny unproven high-tech".

TGV runs very well, except for the weekly labor dispute that's completely orthogonal to the fact that it's not running in a vacuum.

How you people believe the risk of explosive decompression will solve whatever problems a minority of high speed train systems have is beyond me.

1 bar is acutally not that bad.

Hyoerloop pods are smaller, so less lives are at risk.

And most importantly of all. It's not gonna be explosive, unless you rip the thing open with a bomb and you're right next to it.

Even if you rip the tube in half, the air is going to take some time to fill it up. A few miles from the hole there will be a much more harmless pressure gradient already.

Sure it might shake your capsule, but you're not gonna get shredded.

>And most importantly of all. It's not gonna be explosive, unless you rip the thing open with a bomb and you're right next to it.

And that's precisely what makes it unworkable with regards to safety. Burying the tubes is an option but by that point you would be looking at rail instead.

Ok so somebody attacks it, with explosives. And manages to time them well enough that they go off right next to a cart, or inside one.

They will get 1 passenger car. Which is what? 8 People?

The tube will slowly fill with air, everybody else evacuates safely at a slower speed.

I'm willing to bet on those odds. Compared to the number of car accidents, it'a a no brainer.

> 1 bar is acutally not that bad.

That does not mean anything.

The weight of a 1 cm² column of air is 1 kg. So 1 ton for 1 m². Let's say a square profile hyperloop tunnel 1 m high decompressing will be the equivalent of dropping one tonne for each meter. The shock will propagate at the speed of sound.

You neglect the friction that even laminar flow would experience while flowing into the tube.

If you make small fins at the top of the tube you can even exagerate the effect and introduce turbulence, which further reduces flow rate.

There is a reason why pipelines need pumping stations every couple miles, to keep the fluid at speed.

Sure the wave might propagate at the speed of sound but that doesn't matter much if its pressure gradient is slow enough.

https://en.m.wikipedia.org/wiki/Hagen–Poiseuille_equation#Po...

Edit: To give a good analogy: Electricity also travels at the speed of light through a cable, however due to resistance you will get different current reading depending on where you measure.

Edit2: I reread your argument, and there is also a flaw in your energy calculation. The energy per time is not proportional to the length of the tunnel, but proportional to the size of the inlet.

Your calculation would be if you remove the entire top of the tunnel at once, over it's entire length.

Turbulence in a near vacuum? I'm more than rusty on my statistical mechanics, but what you describe will only matter long after most of the potential energy has done its damage.

Don't look at the wavefront of the first 100m, the pod there is gone anyways in case of a bomb.

Turbulence on the phase of what comes after the near vacuum.

The plug essentially plugs itself from resticted airflow.

If you want to be extra safe you could even add small tesla valve style fins around the border of the tube, to increase turbulence.

What about what's right outside the damn pipe? Or do you plan on laying said pipes only in the desert?

Pressure wave moves at speed of sound. Lightweigh capsule gets destroyed just by derailing at 500mph. Any bad event will put down ALL trains on track. And rescuing people trapped in 100 miles long steel pipe...

Hyperloop is a nightmare!

See my above reply. Speed of sound doesn't mean much if there is a constantly softening gradient.

https://en.m.wikipedia.org/wiki/Hagen%E2%80%93Poiseuille_equ...

Edit: To give a good analogy: Electricity also travels at the speed of light through a cable, however due to resistance you will get different current reading depending on where you measure.

The Shinkansen is an existence proof that high speed rail can be fast, comfortable, and always on time (and while by no means cheap, affordable for many). I often wonder what Japan did uniquely that makes this possible.

But building trains that run in a vacuum doesn't appear to be a necessary attribute of running a successful high speed rail service.

The was a CCC talk about why the Deutsche Bahn has so many delays: [1].

It turns out that East Germany has great on-time performance, probably because all the infrastructure is new. But overall, Germany has to invest more in upkeep of its existing infrastructure.

1. https://youtu.be/AGCmPLWZKd8

That's a problem of the ICE not HSR as a whole. I never had issues with the TGV or shinkansen.

DB is a joke in general. Any time I travel through Germany there's been constant delays.

I too have bad experience traveling Basel - Cologne route with DB. At this point, I switched to flying or renting a car. The thought of traveling with DB gives me shudders.

Instead of hyperloop, why not Japan SCMagLev? That things is actually getting build for commercial operation right now.

I think the point is that if your country can’t make it work, perhaps you should look at how the French and the Japanese do it and copy them.

It's actually more like billions of $ per mile. How anyone looks at a figure like '1 billion per single mile of tunnel' and thinks it's feasible to run utopian pie in the sky experiments like hyperloop across hundreds of miles is beyond me.

HS2, under construction in the UK, is 155 miles long, with a total budget of £55bn but currently expected to cost well over £80bn - some estimates have it as north of £100bn. $1bn per mile is in the same ballpark.

In the UK everything costs twice as much as originally estimated and takes twice as much time. I hope I will live to experience the 330km/h (operating speed of UK's HS2)

According to the original hyperloop paper (https://www.tesla.com/sites/default/files/blog_images/hyperl...) it is meant to be able to help with infrastructure costs. The main difference is that it is all much lighter than conventional rail. This means that you can build it on pylons over the ground, and continuously extrude carbon fibre to build the tunnel. In theory both land costs and construction costs should be much lower.

That was the original paper though, I don't know how much of that is considered realistic now.

But the bulk of the infrastructure cost difference of the original hyperloop paper vs. the CAHSR project it was offered against wasn't due to use of pylons, it was due to the fact that the proposed hyperloop didn't serve communities between the termini at all, and didn't come anywhere near the urban centers at either terminus.

This, of course, also made it useless for literally any purpose.

The Shinkansen is built on pylons in many places especially where it goes over low-lying farmland and through cities, so I'm not sure this is a good reason.

This!

Pylons for the win!

Reply on news.ycombinator.com