Hacker News new | past | comments | ask | show | jobs | submit login
Gotthard Base Tunnel (wikipedia.org)
97 points by lelf on Oct 28, 2015 | hide | past | favorite | 48 comments

It looks like the trains through this tunnel will be all-electric. I wonder why they went with a 15kV power system instead of a modern 25kV system?



Perhaps for compatibility with the rest of the Swiss rail network?

You're right. There don't appear to be any 25 kV rail systems in Switzerland.


The software control SCADA system for the Gotthard Base Tunnel is called WinCC OA, made by ETM Siemens. Full disclosure: I did a summer job for them in Eisenstadt, Austria in 2012, and my dad uses the same software at CERN.

ETM made software called PVSS, and it was renamed to WinCC OA when ETM got bought by Siemens. The previous Siemens software, WinCC, was the victim of the Stuxnet worm. Hopefully ETM's is a bit more secure!

This is public knowledge, not insider information: http://www.etm.at/index_e.asp?sname=div_nw_0209_gbt_e.asp

There's a (kind of cheesy) documentary about the construction of the tunnel: https://www.youtube.com/watch?v=QaTN_R1b00I

I skipped through some of the slower-paced sections, and mostly just gawked at the machinery.

Thanks for the link. Cheesy or no, I love these kinds of docs, wish there were more of them out there.

Gotthard Tunnel

The Gotthard Base Tunnel (GBT) is a railway tunnel through the Alps in Switzerland expected to open on 2 June 2016.

- Length: 35.4 mi

- Cost: $10.3 billion

- Construction: 1996 - 2016 (est)

Bay Bridge

The San Francisco–Oakland Bay Bridge (known locally as the Bay Bridge) is a complex of bridges spanning San Francisco Bay in California.

- Length: 4.46 miles

- Cost: $6.5 billion

- Construction: 2002 - 2013 (est)

The GBT is 94 miles, right? I think the Swiss achieve these costs through economies of scale. There are tunnels throughout the country so they can achieve greater efficiencies at 5-10x of those in specific metro areas in the US. I remember when they were debating a tunnel under Tysons Corner in Virginia, the largest office park in the Washington-Baltimore area, the estimate came in at $800 miles for 4 miles and was considered cost-prohibitive. It'd be great to bundle a number of these projects together to achieve the efficiencies of scale.

"Economies of scale" when tunneling through solid rock?

I would naively expect running a tunnel boring machine 1 mile to cost a lot more per mile than running it 2 miles.

FWIW, a surprising amount of the base tunnel was blasted by dynamite, as the geology didn't allow use of TBMs everywhere. (And they hit more cases where they had to blast it than they expected.)

Construction is expensive. Even the bargain priced tunnel is about $65.000 a foot or $200.000 per meter. And unlike the bridges it just handles one vehicle width in each direction.

How does average construction worker pay in Bay Area compare to Germany/Switzerland?

Why Germany? The Gotthard Basis Tunnel is in Switzerland and worker pay in Switzerland is considerably higher in Switzerland than in Germany and also probably the US

while i'm sure the issue with the costs here and there isn't labor/salary, googled out of curiosity - for welder in US it is $30-50K/year avg, in Germany - the government mandated minimum in non-East Germany is something like 14euro/hour, in the East - 11 euro/hour.

I also would be very surprised if the Germany/Switzerland were "outsourcing" the job to Chinese companies like the US does with big infrastructure projects. I mean everybody who ever dealt with outsourcing up close knows what in reality it is completely opposite to "cost savings", instead it is a way to "digest" even more money than it would be otherwise.

Why are 2 comments putting Germany/Switzerland? This tunnel is entirely Swiss and has nothing to do with Germany.

Also, comparing salaries of those countries is just ridiculous, as Swiss salaries can be 2-3x higher on average than the German ones.

The minimum wage that can be paid for a tunnel hand in San Francisco is around $34/hr plus overtime and benefits. See http://www.dir.ca.gov/oprl/pwd/Determinations/Northern/NC-02...

You have to remember that tunnel jobs are almost always run by a municipality which means that they're almost always union. I would imagine it is similar in Switzerland.

Tunneling is a niche industry and commands a high price.

We don't know that, because the government is controlling the price.

And the bay bridge has some serious defects already.

Another very long trans-alpine railway tunnel in construction 250 km to the east: https://en.wikipedia.org/wiki/Brenner_Base_Tunnel It's nowhere near completion, though.

Length: 34 mi

Cost: $11.5 billion

Construction: 2007 - 2025 (est)

I remember when they started this project. This is seriously maybe the greatest work of engineering on Earth.

Now only if the US can match this and finally get that bullet train network up and running...

Why is this "seriously maybe the greatest work of engineering on Earth"?

It looks like straight forward rock tunneling to me. The alps have good rock that the industry has a lot of experience mining through. Don't get me wrong, the length is impressive but the diameter (10m) is not crazy nor is the ground bad.

What makes this project incredible is how it is dug at the base of the mountain, instead of higher up. That makes it the deepest tunnel, dug under 2,500m of rock at its max. When you have almost two miles of mountain weighing down on you… pressure is intense. The rock has different compositions along the way, some parts have to be dug with boring machines, some parts with explosives.

The project has been in construction for 20 years. For comparison: the Golden Gate Bridge took 4 years. The Hoover Dam, 5 years. The Burj Khalifa, 5 years.

Two miles of mountain does not "weigh down on you". As I said earlier, the alps are known to have very good rock. What this means is that it is, essentially, self supporting. You are not designing for two miles of overburden pressure, you are designing for maybe a couple diameter's worth of overburden.

If you read my comment history you'll realize that I work in the tunneling industry. I have designed, inspected, or helped construct large diameter hard rock tunnels, small diameter soil tunnels, and everything in between.

There are deeper mines than that. The long time the project is taking is also an indicator that it is mostly about repeating a certain set of tasks over and over and over again. Certainly there were big challenges, but I imagine they were more of a project planning and management nature than pure engineering problems.

It is quite impressive nontheless.

> The project has been in construction for 20 years.

And, as impressive, it's still on schedule and on budget.

Those are bad comparisons (at least the dam and the bridge) because projects of that scale would be nowhere near that fast in the US because of environmental impact studies, NIMBYism, and just improved safety conditions of the labor itself (tons of people died in the construction of the dam).

Originally, there was going to be a station partway through the tunnel. It would've been 800 meters underground - the deepest in the world. It was scrapped as uneconomical.


The station was actually built, but only as an emergency facility and to help with construction. It's reachable from the surface, but requires a horizontal trip on a mine train and then 800m vertical on a mine hoist.

Stopping trains there interferes with traffic too much.

> it will be the world's longest and deepest traffic tunnel

yet the linked list of the worlds longest tunnels lists two longer metro tunnels in china

That list also mentions a "Longest railway tunnel excluding urban metro lines with intermediate stations", so I would guess it is in that category.

Makes some sense; a metro tunnel with stations every kilometer or so will not feel like a single tunnel, even if it was engineered as one.

Hmm. Maybe they're drawing a distinction between metro/subway and normal (freight) trains?

TFA doesn't mention whether or how far the project is over budget or behind its original schedule. Does that mean it is on schedule and on budget? If so that's amazing.

I'm used to constantly hearing stories like the "Big Dig" here in the US, sometimes it seems like every non-trivial civil works project fails massively fail on both criteria.

Originally, the cost of the complete AlpTransit project (of which the Gotthard is part of) was estimated to be about 14.5 billion Swiss Francs (CHF), in 1998. It ended up costing about 18.7 billion CHF, so quite a bit more. The whole project changed quite a bit in the mean time, though (e.g. added security and environmental measures).

source (in German): https://de.wikipedia.org/wiki/Neue_Eisenbahn-Alpentransversa...

That's only 30% over budget. Coming from a country where civil projects are often used to "vanish" money, I am very impressed.

Also coming from software engineering. 30% over budget is just impressive. I would love to be capable of such accurate planning.

"The new Alpine railroad (NLFA) will cost 300M CHF less than planned, because the digging and concreting of the Gothard tunnel have cost less than expected.

The final bill is expected to be between 23 and 23.5 billions CHF instead of 24 millions CHF."


The shocking thing to me is that they are completing the project AHEAD of schedule!

Another awesomely engineered route to avoid the Teufelsbrücke. https://en.wikipedia.org/wiki/Teufelsbr%C3%BCcke

I've sometimes wondered if a Reno to somewhere in western Sierras base tunnel would make sense in the US. Probably not nearly as much of a bottleneck to justify.

I think self-driving vehicles will make projects like this even rarer. The two benefits this project achieves are:

1. Safer travel on rails instead of trucks (accidents & environmental)

2. Faster transport between regions (saving 1 - 1.5 hrs)

Safety will become almost a non-concern with self driving cars. Work still needs to be done to minimize environmental impact, but electric cars are coming on strong.

Faster transport times will be a tougher sell when we are actually free to be productive when traveling. With an automated driving system, vehicles become offices/lounges.

Self driving cars don't completely address these concerns, but they make the $10 billion price tag hard to justify. I'm not saying this project isn't justified, I just love examining the impact self-operating vehicles will have on our future.

Not so much in this particular case — the purpose of the tunnel is to reduce heavy truck traffic through environmentally sensitive areas. The tunnel is explicitly built to satisfy a goal of transforming truck freight to rail freight. The fast passenger time is only listed as a side benefit.

Not to mention electric is not viable for long haul heavy trucking (for the foreseeable future).

It kind of depends on how you envison that.

With existing intermodal systems, there's already something somewhat resembling electrified long-haul heavy trucking, though largely in Europe. For various reasons, the US has little electrified freight, and doesn't forsee it for the forseable future, though it's clearly technically feasible.

There are several other possibilities for electrified freight, though battery-EV is almost certainly out of the picture. IIRC BMW have experimented with a battery-powered EV tractor-trailer, but it's used exclusively for local delivery at low speeds (~40 kph IIRC) at a range of ~5-10 km. A very limited pilot.

Self-driving cargo vehicles capable of "training" on highways, with catenary or similar feeds, could offer long-haul separable vehicle cargo capacity. I'm not aware of any serious plans for such a system though, and it would require both massive investment in vehicles and roadways.

Present planning at the US EIA is for natural gas fueled cargo -- both truck and rail.

This tunnel as well as its predecessor is mainly for rail travel. Trains are already far safer than cars (and there is no reason to believe self-driving cars will be safer than trains), and already let you work en route. Also, this tunnel is for freight rail too, and self-driving cars or trucks aren't really an alternative to that either.

Also: productivity in the office > productivity in a train >> productivity as the passenger in a car. I say that as someone who commutes about three hours a day by train and Uber.

Working inside of a self-driving car in the future will be much different from working inside a normal car. Current vehicles are primarily designed around the driving experience. When you don't need a steering wheel, gas/brake pedals, surrounding area visibility, a car's interior could be completely designed. No reason it shouldn't be on par with a private train car. It should certainly surpass regular train seating, since you don't have an aisle going through the center of the vehicle as a design constraint.

Getting rid of the driver doesn't get rid of physics. You can't make the car taller or wider or heavier without impacting fuel economy and stability. You can't get rid of intermittent stops for cross-traffic and pedestrian crossings. And you will never be able to match the smoothness of a heavy rail vehicle with a lightweight road vehicle.

Self-driving cars still need a way to pass that specific mountain range. Right now there is the old rail tunnel, the road tunnel and the mountain pass. The latter usually closes in autumn as do many others in the Alps. The road tunnel has a length of 16.9 km and two lanes. All of this creates a huge bottleneck. So unless someone invents a flying car, tunnels will still need to be built in regions like this.

This route is mostly for freight, not commuting.

I'm sure very few people tried to commute across this section of the Alps by car before this.

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