Hacker News new | past | comments | ask | show | jobs | submit login

Though if they really dug that much tunnel under a city in such a short time, maybe there's an actual innovation in a better way to make subway tunnels?

I think putting cars in these tunnels is silly and that subways are obviously the best answer personally, but I'm also watching two new light rail lines being put in next to my house. It does not seem quick (though they're elevated). It'd be lovely if those tracks were somehow underground faster and cheaper than above ground.

I'm happy enough it's finally happening, but if it could be done economically in a way that reclaimed the land the rail lines are on to make a park or more housing? And without trains driving by my windows on both sides? Maybe the boring machine itself is somehow better than what we dig tunnels now.




Though if they really dug that much tunnel under a city in such a short time, maybe there's an actual innovation in a better way to make subway tunnels?

Don't believe the hype that this is "under a city." It's under a bunch of parking lots, which have little to zero underground utilities or other infrastructure.

It's not like they're burrowing under New York. But it might be good practice.


Boring Co is digging a small tunnel under a single parcel of uniform dirt, which is the easiest type of tunneling project.


If there were an actual innovation in tunneling, I feel like they’d be able to give actual specific metrics to rebut critics who argue that it’s apples to oranges. But AFAIK the bottleneck in tunneling is not a mechanical engineering one.


The tunnel boring in Seattle began right at the edge of downtown, went under a highway and a monumental retaining wall (They covered that wall with surveying targets and monitored it the entire time), then pretty deep under a residential neighbhorhood and a canal.

I think the only time it's moving close to high-rises again is over by the University.

Edit: I believe they went so deep because of the volume of glacial deposits in this area. Otherwise they would have been going through gravel. That made building the stations a bit of a pain (the Beacon Hill station still has chronic problems with water. It smelled like a moldy towel for a long time).


What is the bottleneck


Geotech is a bitch. Soils aren't homogenous and vary a lot in both composition and characteristics. There are multiple types of TBMs that are designed to operate under very specific conditions, for example a soft rock/soil TBM excavates through different mechanisms than a hard rock TBM. One of the big problems in any excavation is you have no idea what's actually down there until you start working in it. The underlying material can differ a lot over very short distances. You can very easily be minding your own business driving the TBM and run into a transition that will straight up mess you up. Combine that with all the previous mentioned stuff and you get a lot of cost overruns and time overruns. But yea, subsurface is a bitch.


Great answer, but I also want to add:

Even if you had ideal conditions or an amazing new TBM that goes through any material, you're still dealing with a machine that scrapes at rocks for miles on end. The parts that do the digging need constant replacement, underground, on the inconvenient side of the boring machine, and even if you did the actual digging faster that'd mean repairing it even more often.


I’m not disagreeing with you (I know nothing about geotechnical), but I was thinking the bottleneck is all the bureaucracy and infrastructure factors, which don’t have mechanically scalable solutions. I’m assuming tunneling through Las Vegas is much different than tunneling in NYC or SF or even LA.


Depends where you're digging. Might be water table; might be existing infrastructure; might just be a certain type of rock.


Judging by the pace of the SF chinatown muni tunnel, the digging happens relatively quickly...the slow parts are the parts before and after the digging.

In SF, they spent years surveying, moving utilities, excavating stations, etc., dug the actual tunnel in (IIRC) a year, and have spent subsequent years finishing all of the other stuff that makes it more than a hole in the ground.


To be fair, _everything_ takes unnaturally long to happen in SF due to all the ways a Very Concerned* Neighbor can tie any project up in approval meetings. https://forms.sfplanning.org/NeighborhoodNotification_InfoPa...

[*] About their property values


Yeah, that’s not a very useful comment. I realize it’s popular to blame NIMBYs for everything bad in SF these days (“NIMBY” is tech for “I’m not getting 100% of what I want, immediately!”), but your comment does nothing to address the substance of the observation: the digging was the fast part of the SF project.

Maybe the whole thing could have been done in half the time without public meetings, but that’s neither here nor there. The work of moving utilities and keeping sinkholes from forming in downtown SF is still incredibly complex, and seems to me to make up the bulk of the schedule.


> Though if they really dug that much tunnel under a city in such a short time, maybe there's an actual innovation in a better way to make subway tunnels?

Or maybe tunneling speed simply isn't the bottleneck so no one ever tried going faster. If it takes 3 months to dig out the tunnel but 6 months to fit it out then you've gained nothing.


They didn't invent any new technologies.

And we understandably should be wary of any innovation in processes.

Because often it is just cutting corners for safety and reliability.


His goal is to somehow run a tunnel boring machine on Mars, right? He'll certainly be testing out new tech if that's the goal. These machines are magnificently heavy at present, aren't they?


Talk about premature optimization…


Have you not been following The Elon Saga?

He’s trying to parlay his life’s work into civilization on Mars. Everything he’s investing time and energy into has uses here, today, but the prize he has his eye on is not here.

Does making a tunnel boring machine lighter or smaller have advantages? Sure, if the machines get used to dig more than one tunnel then a more portable design will begin to develop. But nobody is worried about paying well over $1000/lbs to move the damn thing. And nobody is worried about how big a parachute you need to land it in Martian atmosphere without breaking it.

Knowing Musk, he isn’t spending time on this plan if it does nothing for his dream. Therefore there’s a connection. What might that be? What would digging large holes have to do with sending people to an environment with barely any atmosphere and massive sandstorms? Oh.

I expect solving that problem will take ages. If he starts now he might have some major progress by the time he’s gotten a few missions to Mars. By the time habitat management is a problem he may have something that is pretty crazy but not insane.

I also expect that when spending another $1000/lbs on cutting surfaces represents a net savings, you’ll see him working with some pretty bizarre and exotic materials. What that would be will come as a surprise to me, since I haven’t paid any attention to materials science since about the time they started making synthetic diamond.


No, it's just a lot smaller. A 12-foot-wide tunnel is, shockingly, much cheaper than a 21-foot-wide tunnel. It also has essentially zero capacity, much like a hyperloop. If you want to punch useless holes in the earth, Elon Musk is your man without a doubt.


According to an analysis I saw on youtube somewhere, they were able to find companies that dig similar sized tunnels (Boring Co. is just using off the shelf tunnel digging equipment.) They were running comparable if not a bit slower than similar projects that have been done many times before.


This is the most important factor - the amount of people such a tunnel can move is borderline insignificant, compared to a proper subway with a train.




Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: