Cooling the tube – Engineering heat out of the Underground | Hacker News Reader

Cooling the tube – Engineering heat out of the Underground

ianvisits.co.uk

348 pointsmzehrer157 comments

Subways in NYC are not fun in the summer either. I always assumed it was because when they were designed they didn't consider a future state where air conditioners on the trains dump their heat into the tunnel.

I tried searching for a similar study for NYC but all I found was old articles from years back.

It doesn't look like the MTA shares any measurements of temperature in their data feeds: http://web.mta.info/developers/download.html

Does anyone have ideas on how we could get this sort of data for NYC subways?

there is no way that the NYC Subway's AC's dumping heat into the tunnels contributes in any meaningful way to the temperature inside the stations - the tunnel system is enormous relative to the trains. much more it's simply NYC summers are sweltering and the stations aren't well ventilated at all.

Did you read the linked article? The heat of the London Underground tube tunnels is directly caused by heat released by trains (though primarily from brakes and friction, not AC), and has nothing to do with season. It's swelteringly hot down there in winter.

I don't see why it'd be any different for NYC, where I live.

Many NYC subway tunnels are not 'tubes' and are shallow enough to have surface access. So part of the answer is probably that the NYC subway is a more open system.

The same is true of London (also mentioned in the linked article). The linked article is explicitly only talking about the deep tunnels, which NYC also has.

The point is there are more shallow, cut-and-cover tunnels in NYC than in London. London has entire high-traffic lines that are deep tunnels.

Yeah, NYC is a very different system. The deep tunnels are very well ventilated. Go to a station at the end of a deep section, like High Street and you can feel the wind blowing out because of the forced-air ventilation system. You can also see daylight from the subway trains throughout the system, because it's that close to street level and there are ventilation grates every hundred-ish feet.

For London the passengers contribute 2% of the heat. In a more efficient system they would contribute a larger percentage.

Actually it is related to the air conditioning. Prior to AC, the subway and its platforms were cooler than the outside. Still not pleasant on a 100 degree day, but certainly not as it is today: the platform will be significantly hotter than outside. Just gotta survive long enough to board the train.

https://cityroom.blogs.nytimes.com/2008/08/08/complaints-abo...

The Montreal subway system has a very clever way of somewhat saving energy (and emitting less heat): the section of the track at the station stop is higher than the rest of the track. This means that the train's kinetic energy is converted to/from potential energy whenever the train arrives/leaves the station stops.

The Tube does similar, with uphill approaches to stations and downhill departures.

I was wondering if this was used, especially for cars.

It is used in most hybrid / electric cars. https://en.wikipedia.org/wiki/Regenerative_brake and https://en.wikipedia.org/wiki/Dynamic_braking

I feel like in the long term, this may become a liability compared to regenerative braking and level tracks.

For example the Montreal metro currently only allows a train to leave a station when the platform of the next one is free, limiting the frequency of this very crowded system. With modern signaling, trains can creep up to the ones before them and reduce the time between them down to 40 seconds - but it's more difficult with all those slopes around.

It's also making extending platform lengths or moving/adding stations nearly impossible.

In Australia when it's flat, there were still collisions. Trains are very very heavy. The collision report recommendations had been to keep a long distance between trains. For trains ahead, the coming train would slow down and would come to a stop, if it's probably one train length ahead.

I'm not sure how long the distance is though. 40 seconds is really really hard to estimate with very very heavy trains and no weight sensors. Even AI/ML cannot predict this (re: no weight sensors).

Why would a computer need weight sensors? It knows exactly how hard the electric motor worked when it was accelerating the train and exactly how quickly the train accelerated. It should be able to come up with an entirely usable estimate of the train's weight.

Just as an aside, modern transit trains have weight sensors anyway which adjusts the pressure in the air suspension to make sure the train is exactly level with the platforms.

If there were collisions the signalling was crap. Train intervals below one minute are definitely possible. For example the new Thameslink line in London will have (or already has?) intervals well below one minute.

Trains know their position very precisely, with errors less than 20cm, and their braking characteristics and error margins are well known, especially in tunnels where there are now wet leaves or snow on the tracks. Automatic train operation makes trains stop at precisely the same spot each time (for example to match train doors to gates), to the point where increased track wear becomes a concern.

On peak hour in Paris there is a metro every 2min and on the 7 years I lived there I don't recall any collision.

And according to this Wikipedia list: https://fr.wikipedia.org/wiki/Liste_des_accidents_de_métro

The last collision in Paris metro was in 1981

Related: tube map showing temperatures http://www.gizmodo.co.uk/2017/05/its-official-the-bakerloo-i...

So two lines just barely break 30C for a couple months in the year. Not exactly a crisis.

I believe those are temperatures in the actual tunnels, a passenger's experience in a crowded tube car will be quite a bit warmer!

Underground cars don't have AC/heat? At least in the US those are ubiquitous on subways

The deep level lines don't no. Only the sub surface cut and cover style lines.

The deep level London tube lines would be difficult to fit AC units to given the small size and tight tolerances.

Newer units on the subsurface lines do have AC fitted.

Seems odd... Just replace one seat per car with an air conditioner?

Adding AC to the trains would just make the tunnels even hotter and make the problem worse.

I'm wondering if fitting the station with a wall or glass curtain between the tracks and the platform (with sliding doors to let people in/out) could help mitigate this effect. See the MTR in Hong Kong for an example of what I'm talking about.

Subway system in Seoul also has glass partition between the platform and the tracks.

This offers 2 distinct improvements for passengers.

1. noise/dust mitigation: One can sweep the platform floors all you want, but dust from the track will always get to the platform. By installing the glass partitions, general atmosphere of the platform is just more pleasant.

2. less space for AC to cool: It's impossible to cool ALL of the space of a subway system, from miles of tracks to the platform. By installing the glass partitions between tracks and platforms, AC at the subway stations only has to cool the platform.

I heard NYC subway engineers say they can't install the glass partitions because the subways are not capable of stopping at a particular spot. This is required for the glass partitions to be installed.

The problem is a lot of the tube stations are on curves, so each would need completely custom doors.

I don't think it would help in the long run. The tunnels would keep getting hotter, and the AC in the trains would have to work harder and harder.

>The deep level London tube lines would be difficult to fit AC units to given the small size and tight tolerances.

Mines quite successfully manage to cool air 2+ miles under ground, so I'm finding it a rather hard to believe that nothing can be done about this.

I think they're referring to cooling of the train cars.

The deeper tube cars running on older lines have serious space constraints to deal with that make integration of things like cooling equipment difficult and expensive. We're talking about trains that have comically small proportions relative to what you would see anywhere else in the world.

That said, last I had read, the deep line cars are supposed to get AC sometime within the next decade or so.

>I think they're referring to cooling of the train cars.

Don't think so. That is a fixable problem. (e.g. convert the space of 1 seat per car for an AC unit).

The fact that everyone is so stumped by the problem says it's the actual tunnels that are overheating. The fact that some lines don't have AC just exasperates this.

You can pull down the inter-car window at the end of the cars. TThis is why the messy hair style is quite popular in London.

In the cars it gets hotter [1], the humidity is what makes it worse.

1. http://londonist.com/2015/07/itd-be-illegal-to-transport-cat...

Mate, I suggest you take Circle Line during peak hour, and see if the wet sauna is up to your standards :)

> offered a prize of £100,000 to anyone who could come up with fresh ideas

Too late now, but I wonder if they considered district cooling, where e.g. cold water from rivers is used as an alternative to air conditioning. Seems to be used successfully in my hometown of Munich: https://www.swm.de/english/m-fernwaerme/m-fernkaelte.html

> Nobody could think of anything TfL wasn’t already trying, and the prize went unclaimed.

I'd assume they did, especially if it's something that's done elsewhere.

They have used water cooling from rivers in some stations (I can't find the source, but I think it was a previous ianvisits post).

Unfortunately only a couple of stations have an appropriate water supply, and enough free vertical shaft space to fit the pipes.

There are some details on that in the Ian Visits article linked to in another comment.

There was also this, from 1938, although i don't know where they got the water:

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

The article even mentions it!

> Elsewhere, they’ve been using cool ground water to cool some of the stations. An experiment at Victoria station was the first, as water from the River Tyburn was used to cool the air in the station. This was only an experiment, but at Green Park, a permanent version was installed in 2012.

From the article in the top comment :

The Tokyo (and Japan in general) underground/subway is actually quite fresh and amazing in the hot summer. How do they do it? It might be interesting to learn from them and a good question for the Underground of London Engineers.

Singapore's system is also airconditioned. They just engineered it properly when they built it.

In fairness the London Underground is from the 1860s, Singapores if from the 1980s. That said, London's system was designed for coal, so imagine how bad the air must have been back then.

As the article states, there are two distinct types of underground rail in London: the older sub-surface lines, and the newer tunnels. The coal trains were never used in the tunnels.

I'm not sure what the difference is between the two. Aren't they both underground? Here's a video of a steam train going thru the underground to celebrate 150 years of the tube: https://www.youtube.com/watch?v=a3_rxuOFTm8

Generally the difference is how far below the surface they are, and how they are constructed (cut and cover vs. bored)

The sub-surface lines were built by digging trenches and then covering parts of them back up with roads, buildings, etc. This technique is called cut and cover. They're just beneath the surface and large stretches of them are actually open to the air, many were built for coal powered trains. These can be built without tunnelling. They're below-ground but not usually "under ground". The trains are rectangular shapes since that is the traditional shape for a train.

The deep lines (originally called "the tube" although that now refers to the whole system) were built by tunnelling, the trains and tunnels are cylindrical and the trains just fit into them. In many cases the tunnels are 30+ metres underground.

There are differences in loading gauge (things like maximum train height)--you can't run the subsurface trains on the deep tunnels because they're too big. Also, like many subway systems, there's a mixture of surface and underground sections (surface segments primarily on outskirts).

I don't think I get your point. Presumably ventilation is easier to engineer near the surface, but air quality is always going to be bad if oil/coal/coke/smokeless coal are burnt (or are you saying that these fuels weren't used, which would be incorrect).

The claim is that the air in the cut-and-cover lines would be no worse than around a surface coal-burning train. Which could still be pretty bad, though probably not uncomfortably hot.

Contemporary accounts of the Underground during the coal era vividly describe it as uncomfortably hot (and dark).

Long-term heat buildup was known in the design stage to be a problem for Eurotunnel.[1] Huge chilled water plants were built to prevent that from happening. It's a surprise, though, that it would be a problem for the London Tube, which has so many connections to the surface.

[1] http://www.nytimes.com/1991/05/01/business/business-technolo...

How much would it cost to add regenerative braking to the cars?

And instead of ventilation shafts you would probably need active heat pumps

Regenerative braking is already widespread, as mentioned in the article. Unfortunately some of the stock is very old.

Per https://www.ianvisits.co.uk/blog/2017/06/10/cooling-the-tube... there already is regenerative braking on some cars, the issue is that regenerative braking can't happen if there's no train on the same section of DC bus that can accept the power. There needs to be some sort of inverter to sink the higher DC voltage and send it back into the grid.

why not incorporate storage in the car and discharge it when launching from a stop. I am sure there has to be a little onboard storage but it appears they need more if they cannot discharge it. Isn't this ideal for ultra capacitors?

it really reads like the cars must be changed to fix the problem as they are the heat source. so unless an economical means can be found to store/discharge it between stations their only solution is to cool the tube itself.

So isolate the passenger area from the tube area at all stations and force cool air from points where you have easy access to cooling. the air flow would of course move in the direction of trains. can that work?

Adding stuff to trains ("cars"?) increases the weight, which means more energy is needed to move the thing. It also takes up space, which is very limited.

Some storage system near a station (where there is more ventilation) could make sense, but probably costs more than the system mentioned in the IanVisits blog to return power to the city (i.e. convert the DC the trains use back to AC for the city).

I don't think the isolation idea would work. Where does the air for people on trains to breath come from?

Or you can just convert that energy into heat, but do that on the surface. Have large resistor banks on the surface that you connect to the DC grid when the voltage is too high.

That would be acceptable 100 years ago. Not today

Why is moving the place one dumps heat from below ground to above ground not acceptable?

Because you can use it for something else other than heating, like, putting it back into the grid, storing (either battery, flywheel or supercapacitor).

Why? Stopping the trains wastes energy regardless of whether you dump the extra power into the wheels with friction brakes or into a resistor bank. Dynamic brakes on diesel trains already sink the power into resistors.

That would work, but a large flywheel would also be a good solution. You could spin up the flywheel to store energy and if it reaches maximum speed then use the resistors. You'd also need to detect load on the grid and then run the flywheel system in reverse to assist vehicles that are moving.

"run the flywheel system in reverse to assist vehicles that are moving"

Just to be clear, you wouldn't actually run the flywheel in the opposite direction.. You'd take energy out of the flywheel versus putting it in?

Correct.

How about just batteries, or heck, large capacitor banks? How much energy is recovered from a single train braking anyway?

Since insect.sh is on the front page right now, let's try it:

0.5 x 29 tons x (50 mph)^2 to kWh

  0.5 × 29 ton × (50 mi/

h)^2 -> kW·h

   = 2.01233 kW·h

So a Victoria line train at top speed has 2 kWh of kinetic energy.

Given that regenerative braking is ~60% efficient, we're talking about 1.2 kWh. That's really not that much; a typical Tesla battery pack is in the range of 65-100 kWh depending on model. Now, granted, I don't know if you can feed that much energy into a battery pack on the order of ~20 seconds, but using multiple packs would suffice.

So it does seem feasible to do it either by installing battery packs connected to the tracks, or in the cars themselves. I wonder which would be better.

There's no need for storage nor batteries at all, though. It suffices to convert the energy from the DC rails (where the train dumps it) to a bus where every other accelerating train can draw current from -- since there's lots of trains, there'll always be one willing to accept the load.

The only issue is that each third rail section is fed by its own set of independent rectifiers, and the third rail sections are not paralleled together. This means that if there's no other currently-accelerating train on the section of rail that you're on, it's just you and the rectifier substation -- and those rectifiers can't accept your regenerative braking current.

Adding inverter circuitry to the substations would introduce a path for energy flow to go from the DC third rails into the AC grid (opposite of the normal direction of flow, hence the term "reversible substation"). Since the AC grid is connected to all the rectifier substations and since there's many trains on the rail network, there'll always be a source for regenerative braking current that's dumped onto the AC bus.

The question is what's cheaper -- adding inverters to every substation, or adding battery packs on the trains? I don't know enough about the prices of things to be able to guess.

If storage makes sense (if a consumer for the electricity can't be found), battery storage (or flywheel storage) will almost certainly be kept on the wayside and not on trains; that way you don't have to care about weight / volume / heat dissipation concerns. SEPTA has such a system -- a large battery array at a station that captures regenerative braking energy.

I think your mass is for one (empty) car. Multiple result by 8 (or maybe ~10 to account for passengers).

If you can run wires all the way to the surface, you can sink the power into the grid or traditional batteries.

You can do that, but you can also use it to power accelerating trains that live on every other track section / DC bus -- by inverting it back into AC and dumping that onto the AC grid that's used to supply your rectifiers.

I don't understand why lack of space above ground is a hindrance to building new ventilation shafts. Surely these aren't going to be wider than a sidewalk, and in central London the distance between any two roads on a block is rarely more than 50-100 meters.

You'd end up with lots of ventilation grates on the sidewalks on the surface, but that seems like an easy and space efficient solution.

A substantial part of the problem is the sheer age of central London; people have been digging beneath it and piling more buildings on top of it for nearly 2000 years, so given any particular spot in the Tube network, there's every chance that if you try and drill upward from it, you'll hit part of the sewer system, a buried river, someone's wine cellar, something top-secret belonging to the state, a lost graveyard, a plague-pit...

... utilities too. There are probably not many spaces a vertical shaft could go.

You probably don't want a direct vertical grate either, you'll need to catch water, rubbish, people from falling down. An S-bend is going to effect flow.

I wonder if using a different means of regenerative breaking would work, such as hydraulic hybrid.

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

Much more of the power would be preserved, leading to less heat.

Is it feasible to increase the distance into surrounding clay that heat can be conducted? Could metal be used to conduct heat from clay surrounding tunnels to clay that is presumably cooler, further away? I could imagine that if a material exists that insulates electricity but conducts heat very well, the track itself might be useful in transferring heat to cooler clay, making the track colder and cooling tunnels.

...and when _that_ clay heats up?

It's quite incredible that the clay soil is still absorbing heat from the introduction of the tunnels. I heard possibly wrongly that the clay drys out too and ends up insulating the tube lines over decades.

Regenerative breaking sounds like the quickest and cheapest way to address the problems - not that any change would be 'quick' or 'cheap'.

What about a mechanical sling shot system at each sub station?

As the train arrived it gets slowed by a spring or similar system which is then used to propel the train forward once it needs to depart.

Also, what about just slower trains? The heat produced while acccelerating or braking is probably not exactly linear with the speed of the train.

It would probably be simpler to use regenerative braking like a hybrid car. The trains are already electric.

It says in the article they convert some of the heat caused by braking, but I thought regenerative braking meant you also didn't have to brake as much.

Your don't have to engage friction-based brakes as much, yes. And since friction-based brakes work by converting movement to heat, that's why regenerative braking helps with heat.

Many lines already use the simple option of putting the stations higher than the rest of the line, hence converting kinetic energy to potential energy

Interesting article. However I can't recall tube stations being much warmer than outside temperatures in winter (on non-windy days). If that's right then how is the heat dissipated better in winter?

I don't live in London so anyone with regular riding experience please correct me if I'm wrong.

When I was there it was definitely warmer than outside in winter. Don't live there either mind you.

I was wondering that too. If they clay is able to cool off and "reset" in winter then the real story is just that these lines see more use today.

The tube is warm in winter. It would be nice except you are typically wearing a winter coat so it's hard to get it just right.

A given mass of cold air can move more heat than the same amount of warm air.

I never got why the heat in the tube was not being used as a heat source. You could extract the heat and supply it to surrounding buildings at a cost, thereby cooling the tube. The tech is readily available. It would be a win win situation. Plus it'd be be very environmentally friendly.

If they don't have room available to get cool air/water down to the tube, they don't have room available to get hot air/water up, either.

The problem is that the heat isn't a point source, it's diffused across hundreds of kilometers of tunnel. The amount of infrastructure you'd have to build to extract it at scale would simply cost way too much. It's cheaper for any reasonable timeframe to simply continue burning natural gas at the surface for heat than to invest all of this into infrastructure at a very long-term ROI.

Thermodynamics also makes this really hard. The tube is at most 40 degrees celsius. The air outside is perhaps 20 degrees celsius in summer. There's just not enough of a temperature gradient to run any kind of efficient engine.

Natural gas burns at around 2000 degrees celsius.

It could be useful for low-grade thermal applications -- heating (or pre-heating) hot-water supplies, or space heat.

Even simply ducting warm air to street level for outdoor dining (in winter), presuming (against other information, I'm aware) that the ducting could be provided. If you're wasting the heat anyway, put it to some use prior to final venting.

>The future of the cooling the tube project will be judged not so much by how they cool the hot tunnels, but by how they stop tunnels becoming hot in the first place.

That's a very good metaphor for our planet.

"Over 47 million litres water are pumped from the Tube each day" [1]

Use the heat from air in tube to warm the water that is anyways getting pumped up. (Too lazy to do the math to see if this would make any difference)

[1] http://www.telegraph.co.uk/travel/destinations/europe/united...

Is the temperature still rising or have it plateaued?

And how long is it "stored" in the ground? If they closed the tube for a week (I know) would the earth return to 14C?

No. I doubt anything less than a decade would make much of a difference.

It's taken a century or so to raise the temperature in the tunnels by around ten degrees C. That already includes the cooling effect of cold air being pushed into the tunnels for most of the year, balanced by the relatively small number of days a year when the air temp in London is more than 14C.

Without that cooling heat has nowhere much to go. It will radiate out into the air, which will make its way up and out rather slowly. And it will diffuse into the clay/soil around the tunnels, even more slowly.

A fully passive cooling-off period would take years - at least.

The problem isn't impossible to solve. All kinds of active cooling solutions are possible.

The problem is that it's impossible to solve affordably. You effectively have to build a heat exchanger the size of central London, which is never going to be cheap.

The general strategy would be a) introduce less heat and b) extract more. The first might be accomplished through greater efficiencies, though that's limited.

If a major heat component is braking, then locating the additional cooling capacity where breaking is heaviest (presumably on inbound station approaches) might offer advantages -- at the very least this reduces the total treated area for maximum effect.

Given the possiblity of ground-based thermal banking, and the long-term nature of the issue, if any amount of coolant could be circulated through the thermally-affected clay, and made available for seasonal heating needs elsewhere in the city, that might be a net win.

I'm familiar with geothermal energy projects elsewhere (borehole projects in Australia, the Habernero project) where the problem is actually inverted: themal extraction cools the strata around a borehole, over the course of about 40-50 years, to the point that no further useful heat can be extracted.

The thought also occurs that the steel rails themselves are thermally conductive and might be made a part of the cooling system. Not a tremendous radiative surface, but a long conductive length. Poorly placed, that is, low within the tunnel, rather than high, for effective heat extraction though.

Silly question, is the heat so low that it cannot be used for something other than releasing above ground?

It's only about 40 degrees at most. So the answer is basically, yes.

Here is an idea that I sent to the Underground in 2006 when they solicited suggestions from the public.

Add cool to the tunnels, rather than taking heat out.

Build liquid air plants above ground, 2 or 3 floors up in the air so that the heat of the pumps is released above street level and the noise can be kept away from the street. Feed the liquid air into the tube tunnels through insulated pipes which takes up far less volume than air vents. Let gravity bring the liquid air down the pipes. Release the liquid into the tunnels near platforms where the air pump effect of moving trains caused lots of air circulation. Also the car doors open on the platforms.

Since you are liquifying the air, not just the oxygen, it can be safely released anywhere in the tunnels. And if your air intakes are high up you will actually be improving the air quality in the tunnels as well, i.e. cleaner air flows in.

For those who also didn't know about liquid air: "Liquid air is air that has been cooled to very low temperatures (cryogenic temperatures), so that it has condensed into a pale blue mobile liquid. To protect it from room temperature, it must be kept in a vacuum insulated flask." (c) Wiki

Plan sounds great and very expensive.

The flask is for small amounts in a lab. An ore processing plant liquifies gases on an industrial scale and if the temperature is low enough, you can keep it in large vats like any other liquid. Of course I am suggesting something in between where you only make enough to keep the pipes full. Of course it might make sense to make it in big vats overnight and drain them down throughout the day.

This could be tested with a single installation in a single tunnel because it requires no change to the tube system or the cars.

This is a terrible idea. First lets ignore the thermodynamic efficiency losses compared to a regular chilled water cooling system. Liquid air is a potent oxidizer and the Underground has myriad sources of ignition and flammable things like people. Plus the risk of frostbite from exposure is very real. A control failure could kill people and ignite a raging fire underground.

Liquid air is the same as normal air. During its liquid stage it would be entirely enclosed in pipes. It can be safely expanded using the technology used in mine rescue suits like those from Draegerman.

You are confusing it with liquid oxygen which is just as dangerous in its gas form as it is when liquid. Air has some 70+ percent nitrogen in it, whether gas or liquid, and that prevents it from being any more corrosive or flammable than plain air.

That sounds ridiculously expensive. It's much cheaper just to pump cool air through the tunnels starting in the evening, and through the night.

This is the correct answer! Tied to the spot price of power in London, it could be very inexpensive if run only at night (as well as using power delivered back into the grid by trains using regenerative instead of friction braking).

Air use not very good at transferring heat. To have effective air cooling you need to pump lots of air. The current ventilation shafts certainly have their limits and especially night time the noise may become an issue if you put serious jet engines there.

You'd pipe chilled coolant down into The Tube and perform the heat exchange below ground (cooling the air below). The energy intensive chillers would run on the surface.

But first, regenerative braking; you must stop adding more energy into the system before you consider a method to extract the existing thermal energy.

They already use regenerative braking and are running all of their ventilation shafts at full blast all the time (with upgraded fans).

The groundwater system in the article more or less does what you suggest, without the expense of the chillers.

I wonder, how long would they need to shut down the tube before temperatures lowered?

Years, possibly even decades. The problem is that the clay just doesn’t really release the heat - it’s very good at retaining it.

How much heat are we talking about? I'm guessing it wouldn't be enough to use for district heating, the way some industrial waste heat is converted to hot water for homes?

The original source [1] has a graphic at the bottom where it says about 300m kWh per year currently, so about 34MW on average if I didn't miscalculate.

1: https://www.ianvisits.co.uk/blog/2017/06/10/cooling-the-tube...

It's very 'low grade' heat, ie the temperature is rather low. Not useful for electricity generation nor much good at heating.

Also if 50% of the heat is from braking and accelerating at the platform, why not just AC the crap out of each station?

Long term heat build up because of human activities. It may not because of climate change, but this is how humans change climates.

The actual heat released by human activity is not significant in climate change.

What is so "experimental" about the air coolers in the picture? They look like normal A/Cs

It's not so much that the coolers are experimental, but what they're doing with them is an experiment.

Normal air conditioners produce a lot of heat, which is usually ventilated outside. That's more difficult to achieve deep underground.

According to Wikipedia, it's a groundwater based cooling system.

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

The tube did not have powered A/C until fairly recently - it was built with ambient air ventilation only.

And that A/C is only on cut and cover lines (circle etc) not deep level trains.

On deep level trains there is no good place for the trains to dump the heat.

I'd look at how ants or termites cool their nests for inspiration.

By not running trains through them, for starters ;-)

> Most of the tube tunnels have above ground sections, so a hybrid idea is to use air conditioning in the trains when above ground, and while above ground to cool a block of “phase change media”, or water to you and me, into an ice pack. When underground, the heat that would be dumped in the tunnels is absorbed by the ice-pack until it has returned to water.

> Whether this can be viable is still being looked at, bearing in mind that they already struggle to fit air conditioning units into tube trains, finding space for the ice blocks is going to be even more of a headache. And not to forget that the extra weight means more energy needed to drive the trains, driving up running costs.

This can be worked around, do the chilling on the wayside, not on the train! At each station, run chillers that can reject waste heat on the ground -- and chill a nontoxic liquid glycol/water mixture to -40 C. Commercial equipment exists to do this already. Have air/liquid heat exchangers on each EMU, along with glycol storage tanks, a pump, and sensors to keep track of the temperature/volume of the glycol in each tank. On the roof of the EMU, install large-diameter quick-mating liquid connectors, along with fiducial marks. At each station, wayside equipment uses computer vision to locate the fiducials on the EMU, mates with the connectors, does a pressure test to verify the integrity of the connection (squirting glycol is a no-no), pumps out all the warmed glycol (and replaces it with cold stuff), and disconnects. This can be done during the dwell time if the connectors and refill tubing are of large diameter.

Cooling loads are on the order of 50 kilowatts per EMU and inter-station times are on the order of 10 minutes, which means 30MJ per EMU. The ending temperature of the glycol will be on the order of 10C (you need a temperature differential to ensure heat flows from the glycol to the air), its initial temperature will be -40C -- a temperature difference of 50K. Glycol/water mixtures have a specific heat of around 3.2kJ/(kg * K), so we have:

30 MJ = (3.2 kJ / kg * K) * mass * 50 K

leading to a mass on the order of 200 kg, which is quite tolerable for a rail vehicle. The tanks for the glycol can be spread around the car and can be arbitrarily shaped (as long as fluid can be circulated and offloaded) to deal with other constraints. There's no phase changes involved, which makes the heat exchange work non-annoying; there's just liquid glycol and air. EMUs don't need to haul around an air-cooled chiller, all the equipment on the EMU is reliable, does not consume much electricity, and is extremely tolerant of vibration and the harsh environmental conditions aboard a rail vehicle.

If you want to reduce mass further and are willing to accept some more complexity, it might be sensical to use a small chiller on each EMU that uses the glycol for heat-rejection. What does this give you? It means that you can still generate a constant chilled water temperature of 10C, but let the end temperature of the glycol go above 10C -- and more temperature range on the heat storage fluid means more heat energy can be dumped into it. When the glycol temperature gets above 10C, turn on the heat pump to create chilled water at 10C, and reject heat into the glycol (stop before it boils). Liquid/refrigerant heat exchangers are much smaller than air/refrigerant exchangers, so if your compressor isn't obscenely heavy, you can likely save some weight. If you can use the glycol from 10C (when heat won't passively flow from the glycol to the air) to 60C (a reasonable condenser temperature) -- that's another 50K worth of temperature difference, which means our 200kg load of glycol can be cut in half.

More information available from the article's source: https://www.ianvisits.co.uk/blog/2017/06/10/cooling-the-tube...

This is a much better and more thorough read.

These are the first trains that I hear of with no regenerative braking.

> indeed, the use of regenerative braking now converts about half the heat loss back into electricity. However, that can only work where trains are accelerating and braking at the same time, on the same electricity sub-station loop.

Besides that, Tube tracks often rise slightly at stations, so that trains get a small gravity assist to both stopping and starting off again. (Unfortunately it also means that hot air from the tunnels tends to collect there, but you can't win 'em all.)

That is a pretty cool way to implement regenerative breaking that I honestly never thought of (even though the tram in my hometown feeds energy in the grid when driving downwards)

Correct me if i'm wrong, but it seems that regenerative braking is a bit troublesome because it is a third rail direct current system: a "regular" AC system can simply feed power back through the transformers to the power grid, but this is not possible here, so power must be consumed by another train fed by the same rectifier.

It would be much more difficult to feed power back into an AC grid from a regenerative brake. With a DC grid, no synchronization is needed.

That's not true.

1) modern railway is fully IGBT powered. In this case it is trivial to inject current.

2) with DC current you need a substation capable of converting AC to DC (easy: bridge rectifier) but also DC to AC (to given tolerances) which is much more cumbersome.

You're not wrong: However, that [regeneraative braking] can only work where trains are accelerating and braking at the same time, on the same electricity sub-station loop.

Or something like super capacitors to hold the braking energy until the train needs to a accelerate again

The first generation of Munich subway trains (manufactured from 1967-1983) uses resistor banks for braking, so the energy from braking goes right into heat.

Only the later generation B and the new C generation can move brake energy back into the grid.

This is the much better article for this audience. The submitted article is an expansion of tweets summarising this one.

Thanks! We updated the link from http://www.citymetric.com/transport/londons-tube-has-been-ru... to this a while ago.

For one thing I'm surprised they're not using regenerative brakes. It sure will cost some of the profit but refurbishing the trains with these will cut somewhat on that 80% of heat

>> It sure will cost some of the profit

All profits are re-invested with TFL so I don't think cost (within reason) would be a major concern.

They are already using regenerative braking extensively, according to this article

https://www.ianvisits.co.uk/blog/2017/06/10/cooling-the-tube...

> It sure will cost some of the profit

TfL isn't a private organisation, they are a government body. In 2015/16 only half of their costs were covered by ticket sales and other income (advertising, sponsorships, etc). The rest comes from government funding, so taxes.

That's actually mostly not true any more. By next year day to day costs will be 100% self-funded, only capital investment will be funded by the state.

http://www.mayorwatch.co.uk/govt-funding-changes-will-force-...

The surrounding earth acts as a temperature storage buffer, so without ventilation, it will be extremely difficult to remove heat. Instead of the more difficult problem of trying to remove heat, I would rather focus on simply reducing the total heat produced in the lower levels. That can be done by having the lower trains be only for express, so they aren't accelerating and decelerating as much (which is when the heat is produced), with fewer trains in general, and then save the tracks closer to the surface focus for the more heat-producing non-express trains and for the more frequent trains.

On a related subject, the temperature of any cave will remain almost constant at the location's average annual surface temperature. So another option would be to focus on not producing so much heat in the entire city in the first place, to lower the average temperature.

> That can be done by having the lower trains be only for express

You have no idea of the layout of the Underground, obviously. They're not lower trains that follow the same routes - they're entirely different (and very important!) routes which just happen to be a lot lower than some others.

Indeed - and the same line varies in depth across the city. Generally lines are deeper closer to the core, and the Northern Line is anomalously deep due to being built in 1860 when soil physics was not so well understood so extra margin had to be allowed for the river.

Also, there isn't really such a thing as an "express" on the underground.

Thanks, yes apparently I don't have any understanding of London's underground. I was thinking more generally.

The heat from the trains etc. heats the surrounding clay over a period of years. You can add less heat, or you can remove heat from the surrounding clay, or both.

Removing:

Run ventilation on high during winter and keep temps quite low in the tunnels.

Run ventilation on high on cooler nights.

Install cooling tubes in the surrounding clay and cool it directly. Either from above, or from the tunnel itself, a ground-source heat pump (geothermal heat pump) to pull heat from the clay. These can be powered by the cheapest available power, likely solar on sunny days in the future.

Adding less:

Upgrade motors to highest efficiency available. This could halve the waste heat from the motors.

Regenerative braking: if it is too complex to put the power back on the grid, build large "electric kettles" and dump it into a vat of water with resistance heaters. The water vat could be part of a water main so it would be constantly refreshed, and result in slightly warmer water for water users.

Instead of ice in the cars, cool brakes and motors that exceed 100C with water, by boiling the water. This absorbs terrific amounts of heat per kg water.

>cool brakes and motors that exceed 100C with water, by boiling the water.

While this would cool the trains and solve the heat problem over the long term, I doubt most passengers would think that being sprayed with boiling water from arriving trains (and the resulting damp) would improve perceptions of traveling comfort

Reply on news.ycombinator.com