[capkutay]

The tunnel diameter is much smaller. Like 1890s-built London Underground diameter tunnels.

The standard with US subway construction today is massive tunnels with huge, expensive stations. Seems to be a design choice from Musk.

[danielbarla]

I think both of the latter approaches make sense in the context of their era. With classical train / subway-style approaches, it made sense to have a bigger upfront cost, so that you could have fewer (but larger) trains. On the other hand, with today's relatively reliable small-scale electric vehicles and automated driving, I can certainly see lots of smaller pods being perfectly viable.

That said, I don't have any particular insight into the field itself; however I do have a respect for Musk's ability to rephrase the problem just slightly (e.g. landing boosters to save costs) and to turn the whole economics of the situation on its head.

[rsynnott]

> On the other hand, with today's relatively reliable small-scale electric vehicles and automated driving, I can certainly see lots of smaller pods being perfectly viable.

How? These pods take 16 people, and per the article, only one can set off every 30 seconds. This limits capacity to 2k people per hour (even assuming that it manages the promised numbers, and historically Musk stuff doesn't), which is far less than one would expect of a decent bus rapid transport line, never mind an underground train.

[Lambdanaut]

With a 10x increase in tunnel-boring speed, it may be possible to build several parallel tunnels, all serving the same route. In this case, each individual tunnel would be launching 16 people every 30 seconds.

Parallelzation applied to subways.

[jcranmer]

Or, for 10% more cost, you could build a regular subway and move the same amount of people in much less space. You even reduce the truly expensive part of subways (i.e., stations), since you only have to have a mezzanine that covers two tracks rather than 20 tracks.

[badwolf]

Think about it from a redundancy standpoint.

Single big trains get delayed due to any number of reasons - something on the tracks, broke-down train, etc...

With 10 smaller tunnels, they can just be rerouted.

[jcranmer]

Not really. If the other tunnels are essentially full up with other trains, then there's no space to squeeze in the trains from the blocked line. Furthermore, you'd have to have crossovers in place to enable that kind of routing, and those underground switching crossovers are not going to come cheap.

[Robotbeat]

Or just build longer pods. Or link them together like a train. Can still be narrow.

[cyphunk]

small pods introduce the insanity of the highway system into spaces meant for trains. perhaps there are some improvements to be made through automation and scheduling of lanes but it seems possible pods would prove to be significantly less efficient when compared to the traditional use of lanes with large vessels that hundreds of people fit into

[danielbarla]

I guess it depends on just what size vehicles we're talking about. I agree that having tiny one or few person pods would probably not be efficient. But a "10-person cars on rails", type of scenario might not be as insane as you make it sound. Since it neatly side-steps the most complex parts of automated driving, we could have the reality of a well orchestrated fleet of smaller units. These could link up for efficiency on demand (maybe even while in motion), then separate again if they are operating in a complex web. Having a cheaper way of making tunnels feeds back again into the loop. Without doing any rigorous analysis, I can't comment much, but I wouldn't dismiss it offhand.

[ghaff]

So, it would presumably be like loading small trams at a ski area--if they had lots of intermediate stops rather than maybe just one. Yes, those work. Everyone is typically also able-bodied enough to be skiing and you have attendants supervising the loading and unloading.

[lorenzobr]

It's definitely a design choice.

Generally, every city has big transportation hubs - airports, train stations and so on - located at strategic places within the city. Musk's idea is to ditch these hubs and replace them with more frequent and much smaller stations which get you closer to your destination. On the other side if they don't have to move millions of people to the same hub but rather move a much smaller number of people, they can afford to dig smaller tunnels and stations which are way more easy and fast to build.

For instance, in London every time they build a new transportation hub, it takes years. These stations are massive, they literally dig in every direction for several meters.

I don't know if it's going to work but surely there's thinking behind it.

Edit: typo

[phil21]

I think if this tech proves to somehow work (aka actually reduce expenses by at least 10x) the real play becomes PtP tunnels that don't follow the traditional "heavy rail" subway routes, but connect to those stations for transfer.

This also could really only be the long time play anyways - it's the only reason using autonomous battery powered model X's as the "cars" makes much sense.

I think it's an interesting idea, Musk obviously likes his sci-fi. He is basically attempting to implement packet switching for human mass transport vs. the current circuit switching we have.

[jcranmer]

> On the other side if they don't have to move millions of people to the same hub but rather move a much smaller number of people, they can afford to dig smaller tunnels and stations which are way more easy and fast to build.

The actual cost of tunneling itself is generally fairly cheap--somewhere around $50 million / mile. The expensive part is the stations. You can probably save money without having to build mezzanines, but the lower utilization of the tunnel and the greater number of vertical access shafts needed (not to mention the challenges inherent in moving through that very crowded portion of real estate) is probably going to cause cost blowouts compared to subways. Particularly if you design the tunnels to move cars, not people (SOV cars being about the worst use of space possible).

[dullgiulio]

Usually you rate a boring machine by volume/hour so diameter is irrelevant. If Musk is ignoring that to say they are faster, it's snake-oil.

[rorykoehler]

He has specifically said they are making smaller tunnels to dig faster and will use transportation tech that fits in smaller tunnels to take advantage of this efficiency. I think many industries could benefit from this kind of thinking. "Because that's how we've always done it" is a great way to stagnate

[eric-hu]

I think the grandparent is nitpicking the "faster" claim here. To make your comment more specifically address that, Musk isn't claiming to have technology that can bore 10x more volume than competitors. He's reframing the market needs from cubic volume bored to distance bored.

[spacenick88]

But that really isn't the metric you actually care about. The only things that matter is distance tunneled per time and that the resulting tunnel can handle your transport needs. Musks believe is that the transport needs can be handled by smaller diameter tunnels than everyone else is boring. At least if they also built the "trains" and casually looking at subway networks that very well might be true.

[WalterBright]

A smaller tunnel can transport the same volume/time as a larger tunnel, if the transport is correspondingly faster.

[masklinn]

Why would it be? If you can run fast in a small tunnel you can run just as fast in a big one. The size of a tunnel is a factor of what you want to do with it and what you want to run through it. TBM manufacturers can scale down to 3ft if that's what you need.

[Retric]

Train speed is not the limiting factor. Most tunnels have very low utilization in terms of % of time with trains in them.

If you have 4x as many trains but don't load and unload on the same platform you could easily get by with 1/4th as many passengers per train.

[jcranmer]

Fun fact: a subway trunk line is more frequently physically occupied by a train than a highway lane is by a car.

The factors that limit train frequency are station dwell times and switching time. A subway line can generally hit 26TPH, and the top speed of most subways is usually about 70mph, with average speeds generally being in the realm of 30mph. Making trains faster actually reduces capacity; a HSR that goes 220mph is considered to have a capacity of around 4-6TPH. You can also improve throughput by cutting out all branching; Moscow gets about 40TPH as a result, which is about the feasible limit of rail systems.

[spiralx]

The Victoria Line in London runs at 36 trains per hour at peak, one every 100 seconds. Station dwell times are already a limiting factor at this point, and having separate load/unload platforms would provide only a small decrease in dwell time as you'd need to open the unload doors before the loading doors to prevent people just exiting the wrong side.

[WalterBright]

> Why would it be?

Building a new network would not be constrained by existing implementations. For example, you can't just drive the trains faster on existing systems. Everything would have to be redesigned/upgraded to do it - the motors, tracks, track bed, brakes, suspension, safety equipment, schedules, signals, everything.

[masklinn]

> Building a new network would not be constrained by existing implementations.

That doesn't answer the question. Why would the transport be "correspondingly faster" in a smaller tunnel than a bigger one, given both are new digs?

[spiralx]

Air resistance is greater in smaller tunnels though, and aerodynamic drag is the most significant factor on speed when considering underground high-speed trains. This is one of the reasons for building tunnels significantly larger than the size of the train.

[intended]

That’s assuming the current system speeds are being met - as I recall the New York subways are run at lower and lower speeds than capacity to avoid other problems and accidents.

[oska]

If you can still get people through the smaller tunnel you have dug more quickly (because you're using smaller, low-profile vehicles on 'skates') then it's not snake-oil, it's smart.

[dagw]

Don't forget also being able to handle evacuations and emergency situations. A lot of tunnel design isn't just getting a train through, it's getting people out.