Hacker News new | past | comments | ask | show | jobs | submit login
Cargo sous terrain – Underground cargo train in Switzerland (cst.ch)
116 points by sschueller on April 29, 2022 | hide | past | favorite | 128 comments



https://www.cst.ch/en/what-is-cst/ makes it sound as if they're trying to build small-diameter tunnels like the Boring Company, but using the strengths of the format much more cleverly.

The lack of humans means that they can build tunnels with fewer escape routes and almost without regard to ventilation, the centrally coordinated 30km/h speed ought to be cheaper than building for the kinds of speeds one must expect from individual drivers.


I think your ventilation notion is wrong because intensive use of this infrastructure will mean many, many electric motors generating heat. There will also be refrigerated cars, furthering the problem. Having said that, I strongly suspect that they've incorporated the ventilation solution already.

>they're trying to build small-diameter tunnels like the Boring Company, but using the strengths of the format much more cleverly.

Agreed. It's scarcely possible to use tunnels less cleverly than by putting large, single-occupant vehicles through them.


Ventilating well enough to cope with electric motors sounds much less demanding than coping with having both combusion engines and humans in the tunnels.

Not to mention letting the humans escape alive in case a vehicle crashes and burns.


> Ventilating well enough to cope with electric motors sounds much less demanding than coping with having both combusion engines and humans in the tunnel.

You will always have humans in the tunnel. How do you plan to perform inspections, maintenance, and even emergency support when a cart breaks down and blocks your tunnel?


The acceptable conditions for "workers in the tunnel performing regular maintenance" and "a continuous stream of random people" are pretty different. Probably >10x less requirement for fresh air when >100x fewer people are breathing it. Ambient humidity might be lower, and even if it isn't the temperature only has to be workable rather than comfortable or at ambient. Worst case, you could shut everything down and let it cool off before sending people in.

Certain emergencies would also be much easier to handle without people, like nitrogen purging a fire. Realistically I don't think you'd want to shut down the whole tunnel and wait for hours when you need to fix a problem, but it might work if there aren't many problems. It might work for routine maintenance if it's utilized like most packing facilities- deliveries go out in waves, with packing etc in between.


> How do you plan to perform inspections, maintenance, and even emergency support when a cart breaks down and blocks your tunnel?

With pigs of course. Not the ones that go oink though: https://en.wikipedia.org/wiki/Pigging


Yup.

"Must be safe for humans" always shifts the costs an order of magnitude.


The London Underground actually has a heat problem despite using electric propulsion.

Most of the heat in that situation comes from braking.


Wouldn't individual pods break by using the electric motor to charging the battery? I assume the London Underground can't use that method.


It can and does, but that doesn't recover 100% of the brake energy. The Underground will still heat its surroundings, just by less than it did until recently.


Interesting - I have noticed a distinct warmth in a deep station or two (ones with a huge escalator) while traveling there but assumed it was ground heat or something. TIL!


I'm not sure heat is such a problem. The movement of the cargo provides ample ventilation, the surrounding soil has a great thermal mass, and in the worst case it's easier to pump coolant through a pipe than to ensure a breathable, non-toxic atmosphere at all time.


a 95% efficient electric motor is still creating 15x less heat than a diesel engine, and zero carbon dioxide. Traffic will carry out hot air, heat will leave through the tunnel walls, and convection will also carry out heat. At the very least it is a MUCH less relevant issue.


And any cargo container that requires stricter temperature tolerances itself be better insulated and refrigerated. That’s already how standard containers pulled by 18-wheelers operate.


OP's point with refrigerated containers is that they are large sources of heat (since creating a cold volume heats up the external volume by a larger amount), not that they are more vulnerable to heat in the tunnel.


Only if they run their heat pumps inside the tunnel.

If they pre-chill and coast through the tunnel they'd extract heat.

Overall it seems like a non problem with many solutions if any are actually needed.


People act like Boring tunnel are designed to only ever allow single Tesla threw.

They did that because Tesla are available for them quickly.

Do you really think those peope are to stupid to understand that higher occupency can increase threwput?


> https://www.cst.ch/en/what-is-cst/ makes it sound as if they're trying to build small-diameter tunnels like the Boring Company, but using the strengths of the format much more cleverly.

This is ancient technology, literally over a century old.

https://en.wikipedia.org/wiki/London_Post_Office_Railway

No need to name-drop the Boring Company, as it's just a marketing company who bought a COTS tunneler.

Nothing that this company presented was novel. The critical aspect of putting together an railway network through tunnels is a) it's collosal construction and maintenance costs, b) property rights.

The numbers of putting together a tunnel network hardly add up, and ultimately projects fail because there is no way to make the numbers work unless you expect to fleece local and central governments.

It's also quite strange that CST's marketing brochure does not mention at all two of the main solutions to p2p logistics in urban environment: hub and spoke model with the last mile delivered with small electric cars/scooters/bicycles (Amazon already employs this heavily), and drones.


Claiming Boring is only a marketing company is not accurate. Even the first maschine they used was modified. The are are building their own now and they just raised 600M to finish engineering on Prufrock-3 and go into real production.

Just as with other Musk companies, the first version isn't that amazing and people fall over themselves laughting declaring how it will never work and never be practical and how they will never manage to produce them in real numbers.

So its basically like Falcon 1 and Tesla Roaster/Model S.

The leader of Boring is an engineer who was doing well at Tesla.

I think in 5-10 years it will be a very significant comapny.


> Claiming Boring is only a marketing company is not accurate. Even the first maschine they used was modified.

No, stating that the Boring Company is a marketing company that bought a COTS is an accurate statement. The Boring Company didn't invented tunellers digging holes in the ground, and just because it claimed it did modifications to a COTS tuneller (and uncorroborated ones) that changes nothing.

I understand that the Musk Stan phenomenon is a thing, but let's not try to make it look that he invented fire and the wheel.


So any company that buys anything for a starting point and then changes it and evolves from there is not a real company? That is your opinion?

> that changes nothing

And you know this how?

> that he invented fire and the wheel.

I literally said they bought a machine and they are working iteration by iteration to improve it.

Who are you arguing against?


> Just as with other Musk companies, the first version isn't that amazing and people fall over themselves laughting declaring how it will never work and never be practical and how they will never manage to produce them in real numbers.

It's not clear to me what you are saying here. Is it that Musk is clarivoyant, and always selects for lines of business where exponential progress is certain to be made incrementally, or is it that his involvement guarantees that incremental progress will always be made?

As a follow-up to the former: do you think that's the case for Twitter?

As a follow-up to the latter: for how many companies do you believe Musk can simultaneously ensure success?


To throw in my 2c (having worked at Tesla before) - I don't think it is either of these directly. Clearly no one is clairvoyant, however Elon is not one to take the status quo as an answer (eg his first principles mantra). And making a ton of incremental improvements over and over (especially in industries that tend to be risk averse to any changes) can result in huge improvements over time. And it's not like he is directly involved with those improvements. But there is a culture of no BS and fix things if you see it broken, get rid of as much bureaucracy as possible, etc. So if he can instill those cultural values he doesn't need to be directly involved. And if something goes wrong, it can also change on a dime, and go a different direction.


I don't think there's much argument that Tesla and SpaceX have effective corporate cultures, but that doesn't strike me as unique or a guarantee of enterprise success on its own. Nor is "fix broken things" all that new -- that was sort of the basis of the the Toyota Production System / "Lean" revolution in manufacturing that Tesla itself adapted.

My read of GP's comment is that even without any concrete evidence to support a prediction of success (and indeed possibly the opposite), for some reason Musk's involvement alone is enough to be confident that the Boring Company will be relevant in 5-10 years. It's hard for me to understand his logic without writing it off as personality cult stuff.


You don’t think being backed by the world’s richest man, who is a also serially successful entrepreneur, and also an engineer, might make a venture more likely to succeed?


Its not not about what is 'new'. Its about what is consistently effective.

> that was sort of the basis of the the Toyota Production System / "Lean" revolution in manufacturing that Tesla itself adapted.

Yes but Toyota then stuck around for to long on one technology.

> Musk's involvement alone is enough to be confident that the Boring Company will be relevant in 5-10 years. It's hard for me to understand his logic without writing it off as personality cult stuff.

That seems like a strange opinion. If somebody us successful in other conservative high capital business then its a personality cult to suggest that this might be possible again.

And the reason are partly the same. Electrification. So its not even a totally different unique field.

And its not just Musk. I just generally think its a good idea to remove complexity from the tunnel and put it into the pod. One of the big problems with subways is limited numbers and terrible maintenance. You can piggyback of the EV industry.


Both. By selecting conservative industries high capital intensive industries its very likely that incremental improvements can be made. Large companies with large capital investment try to leverage those as long as they can. Musk build his company with consistent improvement as the core strategy.

> As a follow-up to the former: do you think that's the case for Twitter?

I have no idea. Its such a different thing. I don't think its a great use of his time.

> As a follow-up to the latter: for how many companies do you believe Musk can simultaneously ensure success?

I don't think he can ensure it, but it makes it much likely.

But I would say the number is not static. It depends on the problems that exist within the companies. In 2017 during Model 3 ramp it was likely 1 of them, but lucky at that point SpaceX had nailed the most fundamental things in terms of reuse and the company faced on lethal issues.

He he access to lots of great people that he at least partly trust to put into these positions. So that helps, and he pretty good at recruiting young talent.

So I don't really know.


Arguably the key would be picking industries ripe for disruption if you’re willing to look stupid for awhile.

Those would be high-capital low output in general, as low capital or high output have probably already tried many things.

If there’s a Muskette strength it would be identifying those.


> as it's just a marketing company who bought a COTS tunneler.

You mean a marketing company that built their own tunneler, surely. The one they originally bought isn't being used anymore, as far as I know.


They have two now apparently, one in California and one in Texas. But public information on these is sparse, I've not been able to find out much about them. This seems uncharacteristic for an Elon Musk company; Musk and SpaceX talk at length about their rockets for instance. There is a different level of public interest between rockets and boring, which might explain why there is more public information about the rockets. Still, I'm not surprised people are skeptical (or ignorant to the existence) of these bores.


>> There is a different level of public interest between rockets and boring

Or the simple practicalities of the later. I'm a member of the public. What I know of the boring project comes from the youtube videos of the traffic jams in the current Vegas tesla tunnel project. Building an underground tunnel to bypass above-ground traffic problems worked only on the Simpsons (Stonecutter episode). In the real world such tunnels are just another 1-lane road subject to all the same problems as they would on the surface, plus a bunch of bonus problems. It isn't interesting because it isn't practical.


So no subway ever approved anything in terms of traffic?


This is ancient technology, literally over a century old. https://en.wikipedia.org/wiki/London_Post_Office_Railway

See also, the Chicago Tunnel Company: https://en.m.wikipedia.org/wiki/Chicago_Tunnel_Company

According to Wiki, Chicago's started running in 1906, and London's in 1927.


That argument seems parallel to 'Yeah nothing cool about M1 Mac. We've had transistors for decades.'


Well, there's nothing really novel about CST, is there? What's cool is how it puts together elements, IMO.


I mean the M1 exists and you can go buy one.

Paper processors get released all the time.


In the year 90 [i]n the former Roman province of Syria (located in modern day Jordan), researchers are currently studying a sensational canal system. It extends mostly underground over a distance of 106 kilometers (66 miles).

https://www.spiegel.de/international/world/rome-s-tremendous...

https://en.wikipedia.org/wiki/Gadara_Aqueduct


what happens when something goes wrong (or just needs scheduled maintenance) and they have to send a team down to work on it? are the workers going to need scuba gear to breathe?


My guess would be portable ventilation equipment - already commonly used for maintenance access to underground utility infrastructure. They will probably need to provide access to the tunnels from above at reasonably frequent intervals for maintenance access (e.g. to "unstick" disabled vehicles) but since these wouldn't be used for evacuation of more than a small number of people requirements would be very relaxed. The same accesses would serve to run blower ducts when work is being performed, and workers would wear gas monitors. This is all pretty standard for telecoms tunnels, sewers, etc.

Safety evacuation requirements are also easier to meet when you limit tunnel access to a trained workforce under a permit system, compared to a situation where you have the general public in the tunnels. Emergency rebreathers, for example, can become part of your plan. A higher level of safety can be assured for workers since a permit system means that a control center will have positive accountability of everyone in harm's way.


Exactly. Miners have an elaborate safety setup that doesn't have to be anywhere near what passenger tunnels have, because the miners can be trained and react to the situations as necessary, whereas the best you can hope for passengers is they remember the safety briefing and how to exit the vehicle quickly.


Also, miners can be told to wait in a safety area with supplies stocked for the very predictable amount of miners, and so can go a longer without extraction.

Obviously you still want to extract them as quickly as possible but it’s easier to manage.


Why would they? There's nothing consuming the oxygen. You don't need scuba gear to go spelunking. The only issue is removing heat, and you probably don't need much airflow to do that.


There are all sorts of ways for the environment to become toxic; loads of chemical reactions consume oxygen [0], gases leaking from the ground could displace the oxygen eg Xenon gas, etc. Even without those over time maintenance crews will consume the air in an area through both respiration and equipment work.

[0] One relatively common danger to bulk cargo ships is rusting of transported cargo consuming the oxygen in a enclosed space then crews becoming incapacitated when they enter for maintenance or other activities.


Inert gas in an automated subterranean cargo transportation network could be interesting, though. No risk of fire?


It's closer to what Magway is doing in Britain

https://www.magway.com/


This reminds me of when they opened the Gotthard Base Tunnel maybe 5 years ago. It replaced an old mountain railway that connected the North and South ends of the country, and did so at great expense connecting by a straight tunnel that made the route much faster. It takes an incredibly prosperous and stable place to make achieve that increment of improvement. And this feels like more of that same story.

I envy them greatly. The US should be building 5-10 copies of this same system in various regions, connected by a functioning freight rail system for long haul. Our diffuse development pattern and the enormous accompanying expense will be our downfall.


> This reminds me of when they opened the Gotthard Base Tunnel maybe 5 years ago. (...) It takes an incredibly prosperous and stable place to make achieve that increment of improvement. And this feels like more of that same story.

If should be noted that the Gottard Tunnel project was subjected to a national referendum after making it clear the project was astronomically expensive, it would take ages to build, and it was virtually impossible to recoup the investment.


The GBT cost almost 10 billion. Still way less than the 22 Billion the big dig in Boston cost that put the highway under the city. Initial estimate was 2.8 Billion.

Meanwhile the city of Zürich dug a 4.8 km train tunnel and lost (permanently stuck and removal impossible because of possible collapse) a TBM during work almost collapsing the Bahnhofplatzt for just around 2 Billion.

[1] https://en.m.wikipedia.org/wiki/Weinberg_Tunnel

[2] https://www.zvv.ch/zvv/de/ueber-uns/projekte/in-betrieb/durc...


The US has many projects in the $9 billion range, but many of them we never hear about because they're not all that technically interesting, and much of the US is pretty easy to rail over.

For example, who knows or cares about the Alameda corridor? https://www.acta.org even though it transfers tons of freight every single day.


With no disrespect, there's something quite funny about a freight corridor moving tons of freight every day. In the "there are dozens of us!" way.


Heh if their TEUs are mostly laden they’re moving 300,000 tons inside LA every day, which is pretty close to a metric fuckton.


> It takes an incredibly prosperous and stable place to make achieve that increment of improvement.

fwiw, the Italy and Austra (with EU contribution) are building a similar in the same mountain range a few km eastwards

https://en.wikipedia.org/wiki/Brenner_Base_Tunnel


I have very little knowledge of Italy or Austria (pretty little of Switzerland, for that matter), but I gather that Italy can reasonably be understood as a very stable place in exactly the opposite way that the US is stable. Italian places seem very stable, but the wider governance is not enduring, whereas much of the US is held together by a strong national government despite the fact that the places are constantly in flux with most places built cheaply and not for permanence.

Also, it presumably required EU funds to push the project through. My read of GBT and this new CST system is that both are private ventures. That is to say, I take the Swiss projects to be at the tail end of careful analysis, whereas the others may be the result of political desire absent a compelling business case, sort of like trying to make rail transit work in Southern California.


> My read of GBT and this new CST system is that both are private ventures

The Gotthard Base Tunnel was built for the national Swiss railway company, which is fully state-owned. CST seems to be private, although among the largest investors are three state-owned companies (La Poste, Swisscom and the Swiss railway company).

These kind of projects today are always largely funded with public funds in Europe (including outside the EU). It is simply not realistic or desirable to leave them to private entities.


Swiss railways are 51% state-owned, methinks.


100%, only the cargo subsidiary is 35% owned by private logistics companies.


There is another base tunnel being built to the southwest [1].

[1] https://en.wikipedia.org/wiki/Mont_d%27Ambin_Base_Tunnel


It's not clear from your phrasing whether you know this already, but America has an extensive freight rail network already. It's the best in the world in fact. More than 40% of all freight in America uses rail, while in Europe it's less than 20%.

Passenger rail is another matter entirely. Common wisdom about trains in America seems to focus on passenger rail (which is infamously abysmal in America.) But that absolutely does not generalize to freight rail.


Freight volume is one story. The comparison is a lot less generous when you compare freight by value.

US moves a lot of bulk freight by rail that isn’t particularly high value. I remember reading that freight railroads are sometimes opposed to decarbonization plans because a good deal of their volume is coal and oil.


The rail lines are there, regardless of what they're used for. Whatever structural problems exist in this domain aren't addressed by building more more freight rail.


that kind of depends on your definition of "problem."

One thing that is potentially a problem is that freight rail has been so optimized and cost cut to the bare bones that the network is now extremely brittle. This tends to impact passenger rail more than freight rail (passenger rail is legally supposed to have priority, but in practice this has never been enforced and the freight carriers usually just make passenger rail take the hit instead of freight), but this is a problem in major hubs like Chicago.

And I bring up coal and oil, to say that if major changes were to happen in those markets and they dried up, the American freight rail network is not set up to easily pivot to things like fast freight that might take its place, and then we would have a freight rail network that is great for addressing a nonexistent market. Fast freight needs generally straighter, faster lines than what exists today.


Well, if that coal & oil thing dwindles down, there is a new demand known. WATER!


swiss freight rail is the best in the world


This is dumb. Another one of those trains but less-efficient and more costly. Digging tunnels is expensive, even more so across an entire country. Taking out ventilation or escape routes will cut the price, yes - but not enough. And what if something happens inside the tunnel, batteries aren’t famed for their incombustibility…


It's Switzerland, tunnels there always have the secondary use of being usable as bunkers in a war. Also... it's Switzerland, half the country is mountains. These guys know how to build tunnels for cheap, if anyone can scale this to being cheap enough it's them.


Im concurring with you and dissenting with lom. Switzerland is very mountainous that trains laid on the surface will actually be a very very stupid proposition, with active gearing and such and with reduced cargo weight to match. Yes, they have indeed perfected trains on rugged slopes, but only because back then tunneling is a very slow and manual operation that it's not economically and technologically feasible. Lom would be right if the mountains are gentler slopes, but I think lom haven't visited the Alps and appreciate the problem. In this case, the tunnels are actually a far better and more economical solution than overground trains.


Interesting perspective. I have been to the Alps several times, and do realize how limiting it can be. That's the reason the CBT and Gotthard base Tunnel is so important and plays such a central role. The problems start there, upgrading these two massive pieces of infrastructure to support these new types of trains (also note the vendor lock in through novel systems) would be costly and probably not possible without delays for current freight.


I'd normally agree with you, but I don't think you fathom how incredibly starved for space Swiss logistics is. There are densely populated parts of the country where logistics capacity is maxed out, but where there's no more space on roads or to build more warehouses. My understanding is that the CST sprung out of those problems.


Trains are great, but are losing out to trucks, so I think there's a lot of promise in "trains, but small" approaches, even if none of them have taken off so far.

Manual access is a concern, but from what I can tell the vehicles are powered by induction instead of batteries. The cargo might be the bigger fire risk.


> Trains are great, but are losing out to trucks (...)

They really aren't. It's already known that there's a distance and throughpot threshold where roadway and railway make more sense. The main blocker on the ralway side is that infrastructure costs are quite high, and roadway's flexibility always involves trucks handling the last mile.


Building entirely new infrastructure around an entire country for something that can already be achieved by more traditional means is a huge misallocation of money.

> Taken off so far

Other than the Beijing airport train there is no real other maglev running. Even the Virgin Hyperloop has achieved more than this project. And let's not get into what happened to them recently...


> even more so across an entire country.

The website is very clear that the intent is for these tunnels to be dug underneath cities, to relieve traffic burden from road distribution.


Not a train, it doesn't have railway tracks, even on website they call it "vehicle" since it has regular wheels.

Also they PLAN to start operation in 2031, which is so far it might as well never happen.


The Gotthard Base Tunnel was voted on in 1992 and work started in 1999. It was finished 2016. Switzerland plans very long ahead into the future even for local transportation systems. A lot of time is invested in planning which pays out big time in the end.

This project is under evaluation in some Cantons (St. Gallen and Thurgau) for viability and may never be completed.


Gotthard tunnel was just regular tunnel, not trying to reinvent the wheel with new kind of vehicle


Swiss time, not Musk time.

Sorry, couldn't resist.


I’m always amazed every time someone tried really hard to invent new technology that’s just a worse train. I love the overall concept, but damn they’re making it way more complicated than it needs to be.


I recall someone (possibly Adam Something on YouTube complaining about a similar idea) coining the term "Not a Train" to describe concepts such as Hyperloop and this, the idea being that they are (1) futuristic and exiting and (2) because they are different, allow for arguments that the upfront costs will be much lower than trains, whose costs are well studied and known to be high.

Usually these systems are more costly and have lower throughput than trains because: 1. Trains use existing, mature technology and there are plenty existing manufacturers 2. Because trains are coupled together into trains, throughput is higher because there is not a need to space self-driving carriages apart 3. Because trains are coupled together, the expensive bits go in the locomotives and the carriage construction is simpler and cheaper


Trains are great for bulk transport. If you want to ship the daily output of a steel mill or move a military division worth of vehicles, trains are unbeatable. But if you want to ship less than a couple of train cars worth of stuff, they quickly lose out to trucks. That's the market most of these "not a train"s are targeting.

That's why this proposal has tiny cars and even tinier overhead vehicles. It's not so much a train-but-worse, but a truck-but-on-tracks.

Of course we could just put autonomous small vehicles on existing train tracks, but that won't happen in the next two decades because they are incompatible with the existing users of those rails.


Much of US rail is literally trucks on trains - not even containers but actual trailers.

https://www.bnsf.com/ship-with-bnsf/ways-of-shipping/equipme...

This allows a quicker switch to last-mile whilst still moving much of the distance on rail.


The problem with trains is that you need large signalling blocks and distances between trains and blocks for safety reasons - a fully loaded freight train can have a braking distance of well over 1.000 meters, and accelerating and decelerating them takes a lot of time and energy. Also, shunting carriages between trains takes up lots of space for the yards, a lot of time to couple/uncouple and verify the correct operation of the brakes.

The CST system seems to operate at almost zero distance between carriages, which even with the lower speed of 30 km/h should still have a greater total bandwidth than an equivalent-sized two track railway system. Also, it can operate 24/7 unlike rail systems (because of noise complaints and co-existence with passenger trains), which should further increase the available bandwidth.


I can't comment on the Swiss, but in the US if you live next to a dedicated freight line, the train company can and will send trains through at all hours of the night and the only thing you can do about it is find somewhere else to live. CSX, Union Pacific, BNSF, etc could not care less about noise complaints.

If you find yourself living next to tracks, the best you can do is try to drown out the noise like this person did [0]

[0] https://biotinker.dev/posts/seismograph.html


My personal experience was that after a few weeks I just ignore it and sleep through all the noise. The actual train itself was very quiet because of how slow it moved. The real issue was they are required to blow the train horn multiple times at each road crossing. This is extremely loud, since it is a horn.

The only really odd thing I noticed was there was a wall that would squeak as the freight train approached. I suspect this had to do something with the houses construction however.


If the residents in an area get together with their local government and work with the railroad, they can often get the crossings covered to a no-horn Quiet Zone.

The main cost is the double-sided crossing guards necessary, and/or closure of some crossings deemed no longer necessary.

https://railroads.dot.gov/highway-rail-crossing-and-trespass...


This was in Alabama. The local government is generally interested in hookers or embezzling funds.


In Europe, most railways are mixed between freight and passenger trains - in some cases, you have everything from high-speed long distance trains, regional trains, city trains and freight trains on the same set of tracks.

As the railways here belong to the nation states instead of the private freight companies, freight trains are at the bottom of the barrel in priority... and given that it takes a lot of energy to accelerate a fully laden freight train, most operators tend to run freight trains exclusively at night where the amount of stops is minimal.


If you put your house in the acoustic shadow of a tall cinder block wall with a row of trees behind it, you can deflect a great deal of that noise. Build it for shade, just not from the sun.


A conventional train and close headways are not an either-or proposition. Freight trains can (and do in places) operate under computerized moving block systems that allow for headway as short as relative braking distance. Automatic train operation with moving blocks is a relatively mature technology available from a couple of major vendors.

There are certainly advantages to single-unit cars that avoid the need for shunting, but the maintenance cost on these will be tremendously higher on a per-unit basis. Hard to say that it's a clear win.


Once upon a time, motors were incredibly expensive. Not just the motors, but the logistics for the motors. The care and feeding of a coal burning steam engine requires lots of space and people with shovels.

So factories had a single motor, a giant steam engine that fed a complicated web of pulleys and crankshafts.

Electric motors killed those old single motor factories.

The same logic results in a single locomotive pulling a giant train. You need a human per locomotive, keeping the messy diesel in one place helps a lot, diesel engines need a lot of maintenance, etc.

In 2021, electric motors and batteries are cheap and tiny. Trains on a dedicated track don't need humans. Automated swarms of small motors are the future.


Most US trains have a locomotive engineer and a conductor in the lead cab, and the rest of the power units are remotely controlled and unmanned. That's two employees on the train for 3800 tons of freight.

Rail where each car is powered is usually called light rail when used for passengers, but I guess there wouldn't be much stopping light rail being used for freight.

The key now isn't the motors but the remote controlling.


I think the innovation is the removal of manual handling involved in transferring goods to a train or between trucks at a logistics hub. By making the modules small enough that they will ultimately be able to trundle autonomously at low speed from the loading bay at a small factory to the hub and then travel at relatively high speed to the end destination without requiring sophisticated AI to guess what the humans around it will do. The modules can pop then pop up in dense city centre environments and postal workers can distribute parcels the last few hundred meters or the module can trundle autonomously for the last few hundred meters at very low speeds on battery power. Or likewise trundle around and between manufacturing areas at low speeds transporting parts along a supply chain.


That's not what this is though. These carriages will just be going from delivery hub to delivery hub. The deliveries to hubs will be by trucks.

The last mile transport to stores will be by van or bike. And that's a metaphorical last mile; the density of the hubs appears to be pretty low, with e.g. 2-3 hubs in Zürich.

The thing they claim makes this work is bundling all deliveries going to the same receiver (across all suppliers) into the same carriage on the sending end, which then makes it easy to optimize that last mile delivery with small vehicles.

This doesn't seem very compelling.


have you factored in the cost of human labor in your judgement? I mean swiss-level salaries. Thats 4k+ usd / month for retail workers. Thats what makes autonomous infrastructure cheaper in the long run than having anyone loading anything on an already overcrowded train net.


The issue isn’t how automated it is— it’s easy to automate trains and automatically load pallets. Making each car responsible for autonomous navigation in two dimensions and self-loading becomes a much more complicated engineering problem.


More like 7-8k/month if you count true employer's expenses on average employee for persons operating & maintaining complex machineries/connected IT systems.


How is this a worse train?


Because it needs far more complicated and expensive cars for no real benefit.


The benefit is the modularity - trains don't solve the last mile problem, individually router containers on autonomous drives do.


This is not a real last mile solution either. We once had the almost the same level of modularity by detaching cars in motion and letting them coast..

Really this is more about not messing with the existing mix of safety and liability defenses used to keep the current situation on the current tracks. These systems have to be sufficiently different because we aren't going to do a flag day to full automation.


CST isn’t built to solve the last mile problem either—it explicitly shows cargo being delivered to and picked up from central hubs.


trains absolutely solve the last mile problem and there is no better example of their modern usage than in switzerland where your ikea would be supplied by a single boxcar once a week


Reminds me of the 2' wide electric trains that moved cargo under the Chicago loop for many years: https://www.lib.uchicago.edu/collex/exhibits/under-your-feet...


I have a friend of mine who works for the city and he's been able to go into the tunnels. He was working on a big repair project on one because it flooded--they were sealing it and pumping the water out.


This is very strange. Long distance tunnels for small cargoes.

Now, doing that under a city might make sense. Chicago had a system like that.[1] There was dense coverage under downtown Chicago, with most streets tunneled and connections to many buildings. But this was before trucks. It was a solution to the "last mile" problem, getting goods from the railroad stations to business buildings.

The costs of unpacking big trucks or freight cars to put things in small containers would dominate the transportation cost. Manhattan has that problem. There's a whole industry that unloads tractor-trailers in New Jersey and repacks the contents onto smaller trucks for delivery.

[1] https://en.wikipedia.org/wiki/Chicago_Tunnel_Company


This could move packages and mail easily. Those are already containerized, packed and repacked multiple times as they move through the mail system. Larger shipments could be palletized to a form that fits inside these vehicles and be quickly moved too.


Building new trains is always great, but I'm curious why they chose to go for what appears to be a small loading gauge, thus preventing them from using shipping containers.

I was under the impression that containerisation and palletization of cargo leads to huge efficiencies, but surely this setup would require much more labour-intensive (or machine-intensive) loading and unloading than just slapping a shipping container on a rail-car?


For those learning French who might be puzzled by the spelling, the correct french word for something underground is "souterrain" (it is also a noun for an underground tunnel). Probably they chose that spelling to suggest that it is not cargo buried underground, or something illegal (there's that meaning too, like "underground economy": économie souterraine).


Anyone knows the relation with the swissmetro project (see https://en.wikipedia.org/wiki/Swissmetro though it very much reads like an ad) ?


neither will ever happen


This is a much better video than the marketing one they post on the site.

https://www.youtube.com/watch?v=r11X-zMF_pc

(Non-English, use subtitle)


I hope they use it to transport garbage or stuff to recycle with any free capacity. All the diesel trucks that just move garbage around is such a waste of resources.


I'm always skeptical to such big-infra projects for a reason: they are not flexible. We design something today that will be ready after more than a decade on scale hoping to have correctly projected future needs.

Surely trains are very efficient and commercial-only on-rail transports for non perishable/non urgent goods can ALSO run on renewable when there is enough energy also stabilizing the grid and leaving perhaps a side channel for perishable/urgent goods, a thing no other practical means can do BUT they are still not flexible. We design a network hoping that in a decade and for more decades it will be useful to pay it back.

That's why I'm for air and water transports: a plane/chopper/drone a ship/sub/* can go anywhere on air/water, we all needs infra on departure and arrivals places. They cost more in operational terms, but they are flexible and cost far less in infra terms. Development of "flying cars/taxis" and "new cheap ships" seems to prove that such line of thoughts have various advocates. Surely that scale only in a distributed and not that densely populated world, are we heading to a mass genocide perhaps?


just as dumb as all of the other 'definitely not a train' startups, but what makes this one particularly funny is that switzerland already has the best freight rail system in the world.


cooler and more sensible notion of underground transit than that promoted by mr musk!


Is there a way for a train geek to score a ride on a cargo train with the engineer?


Ah I see: poke holes in the ground, much like Swiss cheese.


Even more things consumed won't help with climate change, quite the contrary, otherwise nice little project.


The big competitor to this is autonomous EV trucks. I'm surprised that hasn't got more attention. Seems like a game changer.


Probably because EV trucks would need to go through the current Brenner motorway, that is a very trafficked road, from Wikipedia ( https://en.wikipedia.org/wiki/Brenner_Pass#Motorway ):

The ever-increasing freight and leisure traffic, however, has been causing long traffic jams at busy times even without border enforcements. The Brenner Pass is the only major mountain pass within the area; other nearby alternatives are footpaths across higher mountains at an altitude of above 2,000 metres (6,600 feet). As a result, air and noise pollution have generated heavy debate in regional and European politics. As of 2004, about 1.8 million trucks crossed the Europa Bridge per year.


There's an alternative to only on motorways and never on motorways: sometimes on motorways.

90% of the time you can use ordinary roads, with autonomous driving acting like a normal truck, giving the meat bags plenty of space. The most congested 10% you can build restricted access paths only for your smart trucks where they can act more like a train, with minimal spacing, increasing the traffic capacity of the road/tunnel/bridge/whatever so the extra cost of constructing it pays off sooner and/or it's cheaper in the first place because it's not built for drivers so doesn't need to be as safe.


One goal of a project like this is to reduce the number of trucks on the road, which reduce quality of life for people living near and using the roads, regardless of what powers them.


There's already a big push in Switzerland (And I thik across the whole of Europe) to get cargo off roads and onto rail. EV trucks still take up room on roads and cause congestion. https://www.bav.admin.ch/bav/en/home/modes-of-transport/rail...


With ev trucks you lose tons of freight capacity to the battery since the maximum weight of the truck is limited by regulations


That's really not a factor. The most popular semi truck in America, the Cascadia, is rated for 52,000 lbs GVWR. Kenworth T680 has a 64,000 lbs GVWR. Both far short of the 80,000 lbs limit.

The eCascadia has a 230 mile range on a 475 kWh battery. If you make that 1.5 MWh, it'll handle essentially any trip a single driver can make before his workday is over. A 1.5 MWh battery works out to ~17,500 lbs. Subtract the engine, exhaust, and fuel weight from those trucks and the majority of tractor trailers on the road will still be under the legal weight limit.

The largest factor is the lack of charging (detouring to hit a charger is very costly) and the cost of batteries that size.


If there's one thing I would absolutely require a Level 5 Full Driving Automation for, it's for freight.


I'd assume the opposite.

Run it at night, at economically efficient speeds, in convoys, on motorways and you get 80% benefits with 20% of the effort. No cyclists, roundabouts, driveways, parking, children, back roads, routing you've got less other traffic, could modify the road itself, run more smaller trucks, EV versus diesel, no tired human living away from home to worry about. Use Bus Rapid Transit corridors at off-peak times.

Feels like something thats going to go from stupid to standard very quickly once some tipping point gets hit (probably based on battery availability).



My thought as well, trucks will always be more flexible, using existing road infra. It’s hard to see how tunnelling is a better idea, cool as it sounds.




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

Search: