[ChuckMcM]

Would love to see how they costed that out. Given that I don't know of a single engineering company that can build straight up light rail track for at $2.5M/mile even if you give them the land for free.

Edit: add this update --

Interestingly this link: http://www.freerepublic.com/focus/f-news/843626/posts claims that the ET3 bid was 1.2B$ which is $12.5M/mile which is a lot more credible (still low though since the surrounding infrastructure to keep the tube evacuated, the mag lev stuff, etc all add cost over regular fused rail electrified service (which California is considering for its fast rail) and that is looking closer to $25M/mile in the current state of the art)

[Eliezer]

How the hell does light rail track cost $2.5M/mile? How did people build railroad tracks back in the days of horses and buggys?

[ChuckMcM]

That is a great question, I wish I could reasonably answer it. I'm a big fan of rail, and while San Jose was celebrating their first 4 miles of light rail (in just 10 years from the start of the project!) I was looking that the marker in the Sacramento Rail museum (recommended) commemorating the laying of 10 miles of the Union Pacific rail line in one day.

Part of it is that some folks really hate rail, so much so that they will continually sue anyone who is working to build it. They will argue wildlife endangerment, habitat destruction, cancer risk, suicide risk, traffic risk, earthquake/disaster risk, global epidemic risk, job preservation/creation/destruction risk, you name it. Anything to get back in court and have a judge temporarily suspend work. Because you hire someone to work on a project, and they can't because of some court order, you still have to pay them. So what happens is these projects have 'burn rates' (which is the cost of renting equipment (or depreciating it if you own it) and labor and materials (some of which degrade over time if not used)) and then you have 'able to work' days. Actual work days might be 90 for a mile of track, but time actually passed is like a year. So the other 275 days people sat on their hands while expensive lawyers argued to get work restarted.

Its one of the things I bring up at town hall meetings with politicians. The tax payer, and the 'NIMBY' [1] folks, fight a very asymmetric kind of warfare. No court challenges until funds are committed, and then six. You need look no further than the maneuvering around the California High Speed Rail project to see it play out in all of its ugliness.

[1] NIMBY -acronym Not In My Back Yard for people who are opposed to any new infrastructure near where they live.

[paulsutter]

Haven't you just proven that your initial post above is irrelevant to an undersea rail system?

I'm not saying an undersea rail system is practical or smart. I'm just asking the relevance of your comparisons to US urban construction costs to begin with.

As a suggestion, a different tone might make your post seem more useful and less trollish. For example "An undersea tunnel will require serious construction breakthroughs to be practical. As a comparison, today's costs...". That would be a lot more constructive than attacking the people behind the article.

[ChuckMcM]

Great feedback, thanks. The conversation wandered a bit, and while I don't think of my style as trollish I can certainly see how my emotion on rail interferes with my communication.

My initial point is a prima facie argument, the proposal is impractical by inspection. I certainly stand by that assertion, but as part of the supporting argument we've been discussing land based construction with the implicit, albeit not as well supported, stipulation that sea based construction would always be more expensive than land based.

An interesting way to approach the problem would be to outline the design space in terms of operation cost, development cost, and rate of return and see what sort of solutions, if any, might fit inside that box.

[paulsutter]

What conceivable technologies might make it practical someday?

How can we get back to a society that is able to engage in large scale projects again? Look at the rapid low-cost subway construction in China today.

I think you're a super smart guy with really broad interests, experience, and knowledge. I think you could make the impossible happen if you applied yourself to it.

[arohner]

Great response.

>San Jose was celebrating their first 4 miles of light rail (in just 10 years from the start of the project!)

> Actual work days might be 90 for a mile of track, but time actually passed is like a year

So currently, San Jose can build 0.25 mi/year. Optimistically, with no lawsuits, their actual work days could increase to say, 225 days/year, which is 2.5x faster. So we're up to say, 0.66 mi/year, which still sounds way too low. What are the other bottlenecks?

[ChuckMcM]

Well one of my political suggestions to 'even' the playing field was to create a certification date for a project after which no lawsuits could be brought against it. This would give the project planners a way to budget the years of litigation, and once they were certified they could start and continue until they were finished without interruption (except for the usuals, weather, labor disputes, and material shortages).

I've gotten some positive feedback for that but have yet to find someone willing to actually submit it in a bill.

[jamesaguilar]

Isn't this part of why rail is a lot easier to do in Europe? My understanding is that it is a lot harder to sue over stuff like this there. Maybe someone familiar with Europe's rail situation can chime in.

[_delirium]

The process is definitely different at least in the smaller countries. In Denmark, when the Copenhagen metro expansion was agreed on by the government, the plans were passed as a law through the national parliament. Therefore, everything in it became by definition legal, superseding any contrary legislation, unless it violated the constitution.

The U.S. instead tends to work on a model where the legislature passes general rules, and then agencies administer the rules in specific cases. So, for example, a specific rail plan is proposed pursuant to a piece of legislation, but the plan is not itself a piece of legislation superseding others. That leaves it open to all sorts of lawsuits alleging that it didn't comply with the legislation that applies to it.

[theorique]

Passenger rail is excellent in Europe, but rail freight is atrocious.

It's the opposite in the USA.

The Economist had an interesting report on the reasons for the difference: http://www.economist.com/node/16636101

[Someone]

They built awfully bad ones, by today's standards. Rails weren't prestretched and welded together. Instead, they left gaps between rails to cater for temperature differences (http://en.wikipedia.org/wiki/Track_(rail_transport)#Joining_...)

Modern trains also are heavier and accelerate and decelerate faster than the trains you see in westerns. Both mean that they exert greater forces on the rails.

Finally, safety standards are way higher. That means that material must be of higher quality and tested better, both at the factory and after installation.

[rogerbinns]

There is also the issue that the FRA in the US makes trains considerably heavier than equivalents in the rest of the world. http://www.ebbc.org/rail/fra.html

[mostly_harmless]

I feel that most of the cost is for tunneling and the difficulties of working in such a space as a subway system. as for normal railroads, a quick googling shows that the cost is much more conservative, with a 1995 estimate[1] stating about $250,000 per mile to rehabilitate an existing railroad.

[1] http://tacnet.missouri.org/history/railroads/rrcosts.html

[psyklic]

Light rail is above ground, often operating alongside street traffic.

[abraham]

Slave labor and "slave" labor.

[shmulkey18]

Slave labor was too expensive to use because you risked losing your valuable capital. Instead new immigrants were used such as Irish (East Coast) or Chinese (West Coast).

[abraham]

That's the "slave" labor.

[planetguy]

Slave labour was used sometimes, e.g. the Thailand-Burma railway.

http://en.wikipedia.org/wiki/Burma_Railway

Five hundred men died for every mile of track laid, but hey, they had too many prisoners anyway!

[Retric]

A large chunk of the costs of projects like this is building the long flat tunnel/level ground over significant distances. The actual costs of building a moderate vacuum is not that significant. Where this project falls down is in the maintenance and safety side of things not the infrastructure to send the first train down the line.

Don't forget every 10m adds 1atm water pressure. Building a glass enclosed under sea walkway 30m down is not that much harder than building one 20m down and these projects are going to be using glass.

[anamax]

> a glass enclosed under sea walkway 30m down is not that much harder than building one 20m down

It's 50% harder.

What pressure are the passengers exposed to?

I ask because 1 hour @ 4 atmospheres of 80% N2/20% O2 is risking decompression illness. (Yes, 4 - you've got 1 atmosphere at sea level.) You can reduce that risk by increasing the O2 percentage but if you do that too much, you're risking O2 toxicity.

[planetguy]

There are two problems here.

Firstly: As I see it, an evacuated tunnel at 20m is a very different proposition to a non-evacuated tunnel at 30m, even if the pressure difference is the same. Why? Well, your big problem is always going to be leaks. In an air-filled tunnel you can get away with microscopic cracks, no problem; you've got a small direct interface between water and air, and the surface tension of the water is enough to keep the water in place. It's very hard to force water through a really tiny crack. In a vacuum, however, surface tension goes away -- liquid water at an interface with vacuum will boil, and all of a sudden you've got water vapour filling your nice evacuated tube through every microscopic crack in its five thousand mile length. Nasty. Clearly the joined-concrete construction used for the Transbay Tube isn't going to be sufficient.

Secondly: unfortunately the Atlantic Ocean isn't 20m deep, or 30m deep, or even 40m deep like the deepest point of the Transbay Tube. It's several kilometers deep.

[allenlavoie]

Is there some reason that sandwiching air between the evacuated tube and the water wouldn't work? Granted it would add to the cost.

The ocean depth point is interesting. The article says "engineers would tether the tunnel at a fixed depth." I take this to mean tethering to the bottom and relying on buoyancy to keep the tunnel floating at the right depth, which presumably could be 30m or so.

[planetguy]

>Is there some reason that sandwiching air between the evacuated tube and the water wouldn't work?

Yep, because now you've just got your air rushing into the vacuum through the cracks, instead of water.

The ocean depth point is interesting. The article says "engineers would tether the tunnel at a fixed depth." I take this to mean tethering to the bottom and relying on buoyancy to keep the tunnel floating at the right depth, which presumably could be 30m or so.

Even if you could get it to be neutrally bouyant at 30m (realistic estimate? dunno) you're now stuck with a tube, five thousand miles long, floating free in the ocean and anchored only to the bottom. Even neutrally bouyant, there'd have to be huge strains on the joints due to ocean currents, plus presumably some up-and-down forces as the season changes the water temperature and... heck, even a whale headbutting such a flimsy structure sounds like a disaster waiting to happen. And remember, you can't afford to get any imperfections in your tube or it'll leak -- that probably eliminates anything you might have used to build in a bit of flexibility.

[allenlavoie]

>Yep, because now you've just got your air rushing into the vacuum through the cracks, instead of water.

Fair enough. I imagine there's a trade-off between the quality of the tunnel materials and the number of pumps needed along the way, assuming an imperfect vacuum is good enough.

I'd be worried about flexibility too. Skyscrapers do pretty well in high winds and earthquakes, so I'm not completely convinced that a similar effect isn't possible with an underwater tube. Any bending would have to be extremely gradual, though, if there's any hope of shooting something through it at high speeds. I agree that glass isn't going to cut it.

[rfugger]

Some kind of gel between the layers perhaps?