5 ms·

One of Musk's original points was that cost of tunneling isn't linear with diameter--double the size more than quadruples the price, or something like that. But

by wahern 2y ago

One of Musk's original points was that cost of tunneling isn't linear with diameter--double the size more than quadruples the price, or something like that. But then the insistence on using electric cars seems... counter productive. The London and Budapest subways use very narrow tunnels, about the size of Musk's tunnels. Their engineering is considered anachronistic, but given the cost formulas for tunneling one wonders if planners are being short-sighted or stubborn by insisting on larger tunnels. A subway with narrow cars is better than no subway at all. But then there's probably considerations, like handicapped accommodations (i.e. large wheel chairs) that make London or Budapest-sized subway cars a non-starter today, at least in the U.S., notwithstanding that it'd likely still be far cheaper to build a narrower subway and then pay for private, on-demand road transport for those physically unable to use the subway.

tsimionescu 2y ago
And Musk's actual results were a tunnel that was more expensive and slower to build per km than the channel tunnel. And this one isn't built under the sea straddling two countries with entirely different legal systems, but under a tiny patch of desert in a very permissive legal environment. And the Channel tunnel probably carried more people through in its first few months than the LVCC will have transported in its first few decades. It's a monumental flop on every possible level.
- Veedrac 2y ago
  > more expensive and slower to build per km than the channel tunnel Channel Tunnel: ~$22B / 31mi ~ $700m/mi LVCC: $47m / 1.7mi ~ $28m/mi ??
  - orwin 2y ago
    You're correct, but eurotunnel cost was 4.6B in 1985 English pound, which seems to be half of $22B. I think the 22B figure is the total project cost, including the not-tunnel part. Your point is correct though, I don't see where eurotunnel can be cheaper than... Any tunnel in fact. [edit] you could also compare total costs: 22B for the eurotunnel, 106 for the Las Vegas loop. It's around the same difference proportionally.
    - Veedrac 2y ago
      As far as I'm aware LVCC cost $53m, which includes tunnels and stations. Where did you get $106m?
      - orwin 2y ago
        Sorry, I don't find it back, but I remember distinctly reading about 106 millions for 2.2 miles (and not 53 for 1.7). It was quite recent (like two weeks ago). Aren't they opening new tunnels/stations? Maybe the difference in our data is there. It was just pedantery anyway, the cost are still an order of magnitude below the eurotunnel. [edit] weirdly I can't find cost nowhere. Everywhere it is said that it cost LVC 52.5 millions to build the first 3 stations and 1.7 miles, now there is 5 stations and 2.2 miles, they upgraded all the cars and the cost is still reported as 52.5 millions. There is also no report on what cost what either, this is really obfuscated. I don't care about the exact labor/material cost, but I'd like to know if the subterranean station did cost as much as the tunnels? less? How much? What about the outside stations? Did you find anything yourself ? I was never interested in that company, now I find the lack of transparency baffling.
        Veedrac 2y ago
        Thanks. I haven't found better information, I think it's likely all the best details would be buried in formal filings that are more effortful to dig up and dig through. I wouldn't expect detailed breakdowns though, Boring Company is private and isn't likely to want to reveal more of their private finances than contracting requires of them.
Almondsetat 2y ago
I mean, it makes sense since the diameter is 1 dimensional while the boring area grows quadratically...
- floxy 2y ago
  I'm not up to speed on tunnel boring. Do they really grind away the whole cross-sectional area, instead of doing something like a hole-saw? Where you only machine away some of the circumference? https://en.wikipedia.org/wiki/Hole_saw https://en.wikipedia.org/wiki/Hole_saw
  - immibis 2y ago
    No matter what, the amount of material removed is the volume of the tunnel.
  - dhosek 2y ago
    So you cut a one meter-deep circle in a chunk of rock. How do you get that circle removed? The back is still attached to the rock behind it.
    - floxy 2y ago
      Bore deep enough that the weight of the cantilevered piece breaks off. Or use additional leverage to break that chunk off.
      - thecopy 2y ago
        And then falls on the drill, making it stuck and bends the attachment. This is a solved problem: https://en.wikipedia.org/wiki/Tunnel_boring_machine https://en.wikipedia.org/wiki/Tunnel_boring_machine
Arainach 2y ago
>One of Musk's original points was that cost of tunneling isn't linear with diameter--double the size more than quadruples the price Ignoring all the other challenges of scale, this would surely make sense because the area of a circle scales with the square of the diameter...
pornel 2y ago
Glasgow Subway operates in tunnels with a 11ft diameter — smaller than Musk's 12ft tunnels, and they have driverless trains in operation now.
twic 2y ago
Except in a subway systems, the cost of the stations dominates the cost of the tunnels. So a small saving on the tunnelling itself doesn't make much difference. Check out the December 2018 forecast costs for Crossrail [1]. 3676 million for seven stations, plus "station tunnels east", the construction of the station tunnels at Liverpool Street and Moorgate. 1708 million for running tunnels. 439 million for portals and shafts. 956 for tracks and electrics. [1] https://www.architectsjournal.co.uk/news/crossrail-huge-cost-hikes-on-stations-and-tunnels-revealed https://www.architectsjournal.co.uk/news/crossrail-huge-cost...
- wahern 2y ago
  Without a doubt there's been tremendous inflation for the cost of stations, even ignoring the marginal cost of in-fill stations in dense urban areas. Planners and especially politicians demand ever fancier and larger stations. That just speaks to the amount of cost inflation all around. Notice how the review article mentions how "cheap" the Vegas Loop feels. Would they say the same about old stations in places like Boston, or do they apply a different metric because of different expectations? (And to be honest, I know my expectations are different, because we're accustomed to newer things being prettier, at least superficially, especially in the age of disposal products. It's an understandable characterization they made, but it's not very reasonable.) Learning how to build more affordable, utilitarian stations, just like tunneling, is a lost skill. Like any skill, developing it requires iteration and repetition, and over time the output will become more refined. But to achieve frequent iteration will require dramatic reductions in costs.

by wahern 2y ago

tsimionescu 2y ago

And Musk's actual results were a tunnel that was more expensive and slower to build per km than the channel tunnel. And this one isn't built under the sea straddling two countries with entirely different legal systems, but under a tiny patch of desert in a very permissive legal environment. And the Channel tunnel probably carried more people through in its first few months than the LVCC will have transported in its first few decades. It's a monumental flop on every possible level.

Veedrac 2y ago
> more expensive and slower to build per km than the channel tunnel Channel Tunnel: ~$22B / 31mi ~ $700m/mi LVCC: $47m / 1.7mi ~ $28m/mi ??
- orwin 2y ago
  You're correct, but eurotunnel cost was 4.6B in 1985 English pound, which seems to be half of $22B. I think the 22B figure is the total project cost, including the not-tunnel part. Your point is correct though, I don't see where eurotunnel can be cheaper than... Any tunnel in fact. [edit] you could also compare total costs: 22B for the eurotunnel, 106 for the Las Vegas loop. It's around the same difference proportionally.
  - Veedrac 2y ago
    As far as I'm aware LVCC cost $53m, which includes tunnels and stations. Where did you get $106m?
    - orwin 2y ago
      Sorry, I don't find it back, but I remember distinctly reading about 106 millions for 2.2 miles (and not 53 for 1.7). It was quite recent (like two weeks ago). Aren't they opening new tunnels/stations? Maybe the difference in our data is there. It was just pedantery anyway, the cost are still an order of magnitude below the eurotunnel. [edit] weirdly I can't find cost nowhere. Everywhere it is said that it cost LVC 52.5 millions to build the first 3 stations and 1.7 miles, now there is 5 stations and 2.2 miles, they upgraded all the cars and the cost is still reported as 52.5 millions. There is also no report on what cost what either, this is really obfuscated. I don't care about the exact labor/material cost, but I'd like to know if the subterranean station did cost as much as the tunnels? less? How much? What about the outside stations? Did you find anything yourself ? I was never interested in that company, now I find the lack of transparency baffling.
      - Veedrac 2y ago
        Thanks. I haven't found better information, I think it's likely all the best details would be buried in formal filings that are more effortful to dig up and dig through. I wouldn't expect detailed breakdowns though, Boring Company is private and isn't likely to want to reveal more of their private finances than contracting requires of them.

Almondsetat 2y ago

I mean, it makes sense since the diameter is 1 dimensional while the boring area grows quadratically...

floxy 2y ago
I'm not up to speed on tunnel boring. Do they really grind away the whole cross-sectional area, instead of doing something like a hole-saw? Where you only machine away some of the circumference? https://en.wikipedia.org/wiki/Hole_saw https://en.wikipedia.org/wiki/Hole_saw
- immibis 2y ago
  No matter what, the amount of material removed is the volume of the tunnel.
- dhosek 2y ago
  So you cut a one meter-deep circle in a chunk of rock. How do you get that circle removed? The back is still attached to the rock behind it.
  - floxy 2y ago
    Bore deep enough that the weight of the cantilevered piece breaks off. Or use additional leverage to break that chunk off.
    - thecopy 2y ago
      And then falls on the drill, making it stuck and bends the attachment. This is a solved problem: https://en.wikipedia.org/wiki/Tunnel_boring_machine https://en.wikipedia.org/wiki/Tunnel_boring_machine

Arainach 2y ago

>One of Musk's original points was that cost of tunneling isn't linear with diameter--double the size more than quadruples the price Ignoring all the other challenges of scale, this would surely make sense because the area of a circle scales with the square of the diameter...

pornel 2y ago

Glasgow Subway operates in tunnels with a 11ft diameter — smaller than Musk's 12ft tunnels, and they have driverless trains in operation now.

twic 2y ago

Except in a subway systems, the cost of the stations dominates the cost of the tunnels. So a small saving on the tunnelling itself doesn't make much difference. Check out the December 2018 forecast costs for Crossrail [1]. 3676 million for seven stations, plus "station tunnels east", the construction of the station tunnels at Liverpool Street and Moorgate. 1708 million for running tunnels. 439 million for portals and shafts. 956 for tracks and electrics. [1] https://www.architectsjournal.co.uk/news/crossrail-huge-cost-hikes-on-stations-and-tunnels-revealed https://www.architectsjournal.co.uk/news/crossrail-huge-cost...

wahern 2y ago
Without a doubt there's been tremendous inflation for the cost of stations, even ignoring the marginal cost of in-fill stations in dense urban areas. Planners and especially politicians demand ever fancier and larger stations. That just speaks to the amount of cost inflation all around. Notice how the review article mentions how "cheap" the Vegas Loop feels. Would they say the same about old stations in places like Boston, or do they apply a different metric because of different expectations? (And to be honest, I know my expectations are different, because we're accustomed to newer things being prettier, at least superficially, especially in the age of disposal products. It's an understandable characterization they made, but it's not very reasonable.) Learning how to build more affordable, utilitarian stations, just like tunneling, is a lost skill. Like any skill, developing it requires iteration and repetition, and over time the output will become more refined. But to achieve frequent iteration will require dramatic reductions in costs.