Hacker News Reader

Comment by earthscienceman | Hacker News Reader

There are some great questions in here. But what I'm most curious about is how the keep the individual segments from sinking into the seabed at different rates. I can't imagine that the interfaces would remain water tight if any segment sunk by, say, more than 30cm. Can we really be certain about the stability of sediment underwater at the loads that will be applied?

View on HN

Looks like they’re over digging the trench, then partially backfilling with aggregate. So the sections aren’t sat on sediment, but rather a thick bed of gravel type material that will distribute the load over a larger space.

I would also be interested to know what the tunnels effective weight will be once it’s submersed. Given the sections are hollow, the road and train sections will create a buoyancy effect. Possibly enough to offset the total weight of the concrete, given they need to fill ballast tanks in-order to sink the sections. If that’s the case, they can probably adjust the ballast tanks to keep the sections near to neutrally buoyant, and thus apply little pressure to the sediment below.

This makes sense and, although I didn't include it in my comment, the buoyancy was part of my curiosity. While it's clever to take advantage of the buoyancy created by the air cavity, if they're nearly-buoyant that would also mean that the variability in traffic would create a significantly variable (as a percentage) load. I wonder how this is accounted for.

For the record, I don't think I'm "gotcha-ing" the engineers by thinking of these questions. I'm sure there have been answer to these questions since before the project began, it's just that I would like to know the answers.

They also must have a way to seal off segments in case of a water breach? Although, that seems awful given there will be human traffic in the tunnel

See the video for answers:

After putting in a segment, the trench is backfilled with rocky material (over the segment), so that any buoyancy is negated by the fact that it's held down by a large layer of rocks. This layer presumably also prevents most, if not all, leaks from generating instant floods in the tunnel.

Lastly, once completed the project will result in 5 parallel, but mostly independent tunnels (two train tunnels, two highway tunnels, and a service tunnel between the highway tunnels). I presume that the tunnels can be isolated and waterproofed seperately.

Concrete is 2.4 times the density of water, according to Google, steel rebar is about 8. The video mentioned that they were plugged (and thus full of air) and were being given additional lift by the barges so they could float and be towed.

So it appears they will not be buoyant.

I'm not an engineer so I might be missing something, but according to Wikipedia[0]:

> The precast concrete tunnel sections will have a rectangular cross-section that is about 40 metres (130 ft) wide and 10 metres (33 ft) high

and (also mentioned in the video):

> The tunnel will consist of 79 standard elements with a length of 217 metres

If it's a perfect cuboid, this should result in a buoyancy corresponding to ~87,000 tonnes. According to the video, each segment weighs (over) 73,000 tonnes, though this presumably doesn't include other stuff that will be added into the tunnel like road surface material, rail tracks etc.

[0] https://en.wikipedia.org/wiki/Fehmarn_Belt_Fixed_Link#Tunnel...

Here's some info about the seals between sections from https://www.trelleborg.com/en/media/products-and-solutions-n...

    Trelleborg’s Gina gaskets and Omega seals fit between the sectional elements of immersed tunnels to prevent water ingress due to external water pressure. 
    Designed to handle intense transfer of hydrostatic loads and movements between tunnel ends caused by environmental pressures such as seismic activity, soil settlement and temperature effects, Trelleborg’s sealing systems promise a product life expectancy of up to 120 years with little-to-no maintenance.

Not a civil engineer but have done some work in construction. What you're describing is referred to as differential settlement, and can be a problem for any structure (house, skyscraper, bridge, etc). I don't know enough to guess what they're doing here but usually there's some work with the underlying foundation (ie dig and fill back in with something uniform, put piers or piles to reach a more stable surface, etc).

The other things you might worry about is expansion and contraction due to temperature changes, and even just minor size differences in manufacturing (for instance when using moulds to manufacture concrete, those moulds themselves might not all be perfectly the same size or they might change over time). So often in concrete bridges (or even concrete/steel pipes) there will be "joints" designed to absorb changes within some tolerance (eg with pipes, there's typically a rubber gasket that keeps things air/water tight while also allowing some compression/expansion).

at least underwater and covered over by rocks, temperature variation are probably generally less than open air.

While the surrounding seabed can be a nice thermal sink to stabilize temperatures, the aggregate around it could be fairly good insulator, there are tubes in London that have warmed the underground to such an extent that it would take months to cool down. Understandably, lots of the heat in the London tunnels is from braking which shouldn't be occurring in this tunnel.

Point is, I don't think one could model the Fehmarnbelt tunnel as open air or assume that the surrounding material is an infinite heat sink.

https://citymonitor.ai/transport/londons-tube-has-been-runni...

As the segments float when they are filled with air (this is how they're floated out for installation), the only reason they won't float back up when installed is that they're covered with a layer of rocks afterwards. Thus I don't think the load on the seabed will be as great as perhaps you fear - it could be very low if that's what they need to make it work.

This isn't correct. The rocks on top are not to prevent the tunnel segments from resurfacing.

The tunnel segments have ballast tanks which are empty during towing, then are filled with water to increase the density of the segment and sink it to the ocean floor.

You are probably correct - I assumed the ballast tanks were part of the end doors, which were later removed, but re-watching, I think I read too much into that part of the video.

I'm not familiar with the details of tgia project, but i know that for other submerged tunnels they pump grout (kind of liquid cement) under the segments, so both segments on a joint rest on the same chunk of grout. There are probably pipes embedded in the tunnel segments during construction xfor this purpose

Reply on news.ycombinator.com