There's recently been a scandal unfolding in Norway where the parliament has been building a 250 meter long tunnel into its parking garage. Almost exactly 1/100th the length of the Lærdal Tunnel.
So far this project has cost 2.3 billion NOK, or 1.3 billion more than the Lærdal Tunnel. That comes out to 1180 million USD per km.
(To be fair, included in that price is also the renovation of a building external to the parliament).
The highway in question has been a weird-ass pet project since the 60's. It's been routinely denied and shelved many times, but after Stockholm implemented congestion tolls in 2006, it had a resurgence.
Originally it was all aboveground and bridges, but the reason it's now morphed into the longest road tunnel project in Sweden, is because the response to every small NIMBY protest has been "fine, we'll dig it down, but it'll cost more". Note that it's still aboveground in some places, and in something that I'm sure is pure coincidence, those areas are among the least affluent residential areas of Stockholm. Note that everyone living above the tunnels will still get to enjoy years of underground rock blasting, I'm sure that'll be fun.
And since almost the entire highway is now in a tunnel, and since it's going to be the longest in Sweden, the budget has gone up dramatically. How to finance it? Oh, our glorious politicians decided to use money from the congestion tolls. Never fucking mind that the initial promise of those were that the funds were going to be used solely for public transportation projects and improvements - things that would actually solve congestion. But all of that is now put on hold for the next 10-20 years, because this one road eats all the budget.
Yet another criticism is that the highway doesn't really solve the problem, because there's actually very little existing traffic that goes between the points it will connect. Truck transports from the south of Sweden to north of Stockholm take different routes already, they don't go through Stockholm. So you won't get rid of all the heavy traffic that's currently congesting the E4, because most of that is actually going into or out of Stockholm. All the ports of Stockholm are over on the east side of the city. Yes, if you live in the far west suburbs of Stockholm, you now get an easy way of going to the southern suburbs, good for you, but that's going to increase total traffic, not decrease it.
But the most annoying thing about the project is that in this latest round where a ton of challenges forced it underground into tunnels, which resulted in significant increases in cost and time, noone has re-evaluated what the traffic situation in Stockholm is going to be like in 2026 when it's estimated to be completed. (Note: Of course it won't be completed in time or on budget, so 2026 is highly optimistic already). Here on Hacker News there's regularly articles about our glorious new electrically powered autonomous rideshare future of transportation, but this highway project doesn't take any of that into consideration. They're just assuming that traffic will increase like it has the past few decades, they're assuming it's going to be 100% ICE-powered human-driven cars. And yet you have articles like this which heaps praise on the environmental concerns and solutions of yet another ridiculously expensive highway that noone really needs or wants.
The route between Kista and Huddinge is perhaps one of worst traffic route in entire Country. This is because there is HUGE traffic almost 8-9 hours a day on this route. I drive this route twice a month and its horrible. So I don't know who told you that there is very little existing traffic here.
This huge project is a "bypass" and it'll work as any other bypass in other cities around the world. Which means any kind of traffic moving from North to South and vice versa, will bypass entire Stockholm, and right now I see huge amount of traffic every day on this route.
I don't remember the exact numbers, but out of the heavy truck transports currently clogging up Essingeleden, only 2% (Or was it 0.2?) or less would actually benefit from Förbifarten, because there is very little existing truck traffic going from way south to way north of the city.
> right now I see huge amount of traffic every day on this route.
Yes, but most of that is going to places along that route. So you won't move a meaningful amount of existing traffic, you'll just enable new traffic. That's generally good, growing regions need more infrastructure, but my point is that Förbifarten is the wrong way to spend that money. It's a prestige project at this point.
If you had to spend that money on roads, I would have much rather seen Österleden, because completing the highway ring around Stockholm actually makes sense, there is plenty of traffic from the ports to south of the city, and by making the full ring, that traffic would get a choice of direction around the city depending on time of day and commuting patterns. And this holds true for every trip across opposite ends of the city.
The congestion charge backtracking is an egregious example of politicians breaking their promises, definitely agree with this as well.
 = https://vaxer.stockholm.se/projekt/forbifart-stockholm/
In this case, the tunnel is very long and deep, magnifying the unknown territory risk. Also, the bedrock in most of Scandinavia (and, I presume, Stockholm) is both very shallow and mostly granite, which is extremely hard and thus not a good fit for a TBM.
Source: My childhood in Helsinki was punctuated by the periodic sound of blasting rock as they cut out what was supposed to be the new central station. Overall, Helsinki is basically granite Swiss cheese due to parking garages, pedestrian tunnels, basements, bomb shelters, military tunnels, the metro etc and AFAIK all of it has been drilled and blasted, not TBM'd.
There is also a "form factor".
TBM's dig "round" holes.
The cross section of those Stockholm tunnels (I know, I actually worked on a few tenders for them) is very wide and flat.
Generically speaking for a TBM to be competitive you need to have a "minimal" length of tunnel, a TBM has huge - really huge - mounting/assembling (and later disassembling) costs and quite a lot of space at both ends of the tunnel.
Specifically, both (they are two tenders, E302 and E308) tunnels have not a "direct access", there are two "service tunnels" (of a much smaller cross section, rather "steep" and non-linear) that allow access to the main tunnels from the outside, and the outside area available is anyway small, so a TBM could not have been used anyway, even if the cross section wasn't so "flat".
Back to generically speaking, the rules of thumb (approximate, only to give you an idea) for choosing NOT a TBM are:
1) anything shorter than 5-7 Km
2) anything larger than diameter 12 m or not round or simil-round
3) anything in terrains that are too "hard" or too "soft" 
To give you a rough approximation, a (single bore) tunnel made with traditional excavation can have a production of 100-120 m/month, a TBM can usually dig around 400 m/month, so it is much faster.
There are also "mixed mode" approaches, where you (quickly) bore a smaller "pilot hole" in the center of the tunnel by means of a small TBM, usually 3.8 or 4 meters in diameter, so you have a 1:1 representations of the terrains and can (from the small bore) perform any kind of consolidation/impermeabilization etc., then you enlarge the hole by traditional blast and drill.
The much reduced amount of explosives needed for enlarging the hole and a great semplification in the ventilation plant , besides the possibility to always work on "known" terrain make this a very good approach in my experience, and the blast and drill enlargement of the pilot hole is much faster than "full face" excavation, the same rough estimate is around 150-180 m/month.
 or more generally where large variations in the nature of terrain is expected, a TBM for "soft" is different from one for "hard" and viceversa
Drill and blast are the most common by far, especially in mining industry where they are lots of twists and turns, and no-one cares about noise etc.
> Additionally, the original drill, which was said to drill 100 meters per week, broke down after drilling only 18 m (59 ft). The rock was too soft, so the machine could not use it to pull itself forward. The contractor tried to drill traditionally, but had to spend a lot of effort on sealing the water leaks. The contractor went bankrupt and a new contractor, Skanska, was contracted. The new contractor had similar trouble but a better contract that gave compensation for troublesome rock conditions.
TBMs can also bore in soil or sand.
Aside from disturbance and raw feasibility (not sure you can D&B underseas) the main split point is going to be the length of the tunnel. TBMs mean large up-front fixed cost but have a lower price per distance and higher time efficiency, so D&B is more suitable for shorter tunnels (how short will also depend on the diameter you need).
That aside, TBMs are awesome, I got to visit one as a kid (the operator gave tours of the TBM during the few weeks of work prep' before digging actually started in earnest) and it was a great experience, they're impressive bits of machinery & technology.
For long tunnels, it isn’t uncommon to have two TBMs that meet in the middle and are abandoned there (they bore a side tunnel long enough to contain the TBM, and the side tunnel is closed behind them)
For example, see https://untappedcities.com/2017/02/09/the-200-ton-tunnel-bor...:
”Burial of such a large machine was unprecedented in New York City, but is common practice abroad, reported the New York Times in 2011. The Spanish contracting company working on the project deemed it would be more cost effective to leave Seli in place. It was estimated that removal would cost millions and cause delays in existing construction projects. Dr. Michael Horodniceanu, the president of MTA Capital Construction, estimated at the time it would cost $9 million to remove Seli, which had cost somewhere between $6 to $8 million new.”
There is also the setup- carting in the pieces of the TBM and getting them into place, etc.
With Seattle's design/build contract, the construction company owned the TBM. To them it's just a tool whose cost is part of the construction package.
Because a TBM itself is a large capital investment. A small TBM (~7m diameter) is ~10m€, a large TBM probably reach 100m€. This is money you have to pay upfront before the TBM even reaches the dig.
The capital investment of D&B is a (few) jumbos which are cheaper (even large and expensive jumbos are unlikely to breach 2m€ brand new), more flexible (they're also used in mining operations) and less dependent of the manufacturer — thus easier to rent or buy used.
> can't you resell it afterwards, or rent it instead of buying?
The "back end" (gantries) can be sold back to the builder and refurbished, the "front end" (shield and cutter head) is thrown away (some are used as monuments) or recycled: https://www.herrenknecht.com/en/services/global-services/tbm...
That's still expensive as hell, and you've still sunk the money in until the builder has examined and bought back the gantries after the project has ended.
I like how counter-intuitive is this statement.
I used to live in a place where the soil was part of a river delta. It was really easy to drill but impossible to keep the tunnel in place and free of water. And it sometimes just sinks the above-ground facilities: https://www.20minutos.es/fotos/actualidad/24-horas-en-fotos-...
> ..."Drilling and blasting isn't so time-consuming, but sealing the rock from water leakage is. If the rock is very fractured, we need a lot of concrete sealant or reinforcement columns to ensure integrity," said Brantmark, describing the work of the three Atlas Copco Drill machines in use.
Now I live in Stockholm and it's pure rock. These are pictures from the subway: https://www.boredpanda.com/stockholm-metro-art-solna-centrum...
The distance is much more than the channel tunnel, and the population that would use it is much less, but it might still be feasible simply because it could be routed almost entirely through hard granite. In Helsinki a lot of the kind of infrastructure that is placed above ground basically everywhere else is being dug underground simply because the cost of digging is less than the cost of land, and so everything that doesn't need windows gets sunk in the rock. And when you dig, you actually get to recoup some of your expenses by selling the stone you quarry...
Nowadays, one of the main problems of digging in Stockholm is that there are already so much stuff dug out that it is hard to find room for new tunnels!
Regardless, I have no doubt the Swedes did an analysis and weighted the pros and cons.
It may however be involved in the water table, which is always critical.