Cooling the London Underground

A cold and windy evening found me wandering over to the Building Centre for a lecture, ironically enough considering the weather – on how to cool the London Underground. The lecture was being organized by The Engineering Club, as part of the Building Centre’s exhibition on subterranean London.

The lecture was being given by Kevin Payne who is currently the Director, Cooling the Tube Program, Transport for London – and he has held that post for the past 12 months, after several years with London Underground.

A bit of background was offered in the form of an old poster advertising the Bakerloo Line in 1906. The slogan promoting the line, which may raise a few sweaty eyebrows today was the claim “The coolest place in hot weather” – promising a maximum temperature of just 15.5 degrees Celsius (60F).

At the time, this was indeed correct – so what has happened over the past century to leave us gasping for cool air as soon as we emerge from a tube train?

The key issue as it turns out is that the tube traps heat over time and if not vented away it will linger in the London clay which the tubes predominantly tunnel through and build up over time. Indeed, most tube tunnels reach the point where the ambient heat in the clay surrounding the tunnels becomes uncomfortable roughly 20-30 years after the tunnel is completed.

It was interesting to learn that this ambient heat in the clay is not just a problem for cooling the tube, but can cause structural problems as the clay dries out slightly and cracks. It isn’t quite hot enough down there to be baking the London clay into a gigantic terracotta tube, but there have been instances where grouting has been injected into the clay to fill in the cracks before they damage the tube tunnel linings.

So, cooling the tube is not just an issue of customer comfort – it has very real structural needs to try and prevent damage to the network.

Such problems are not as significant in the sub-surface lines, such as the Circle and Metropolitan lines – as they were built with considerable ventilation for the original steam trains. Fortunately, thanks to the ventilation – the new trains which should start appearing next year will come with air cooling systems built into the carriages. This will cool the insides and transfer the heat to the outside where it can be vented away.

What about the deep level trains – can they be cooled by the same method?

Well, it sounds viable – cool the interior of the carriages and dump the heat into the tunnels. For the reason above about the ambient temperature, that is not really viable – but very surprisingly, that is not the main reason they don’t use air conditioning on the deep level trains.

It could actually be incredibly dangerous!

The main problem is caused if a train stops in the tunnel for any length of time. The air conditioning would carry on dumping heat from the carriages into the tunnels, until they eventually “trip out” due to the heat. Then you get a really big problem as the train is now sitting in a pocket of very hot air – and that heat starts flooding back into the carriages. If you think it is hot now, you have no idea how bad it could get if a train got stuck in a tunnel for a period of time.

So, alas – no air conditioning on the deep level tunnels.

A brief slide was shown which displayed the temperatures on the Northern Line between Chalk Farm and Hampstead for the summer of 2006. What this was to show is that the temperature at the surface doesn’t really affect the temperature in the tube – on a day to day basis. Throughout the summer, the tube maintained a fairly constant 30 degrees, while the surface wobbled between 20 degrees at night and as high as 34 degrees on some of the hottest days. Yes, the tube was on two days actually cooler than the surface.

Why is the temperature so constant? Well, it turns out that about 80% of the heat in the tunnels is caused by the trains and services themselves. All those engines, motors, etc are the main problem – and with more trains coming into service all the time to cope with massively rising passenger numbers, things will only get worse. That’s slightly pessimistic, as developments should enable the tube network to run more trains, but for the same amount of energy – so while the best we can hope for is a stabilizing of the temperature caused by the workings of the tube network.

Humans add very little to the temperature of the tube.

OK – so we know that sub surface lines can be – and will be improved, but what can be done to improve the rest of the network?

There are three areas of work – boring but sensible; interesting and innovative; then the blue sky projects.

Over the past 3-4 years, a lot of hidden work has been carried out – and while not the sort of thing to get media headlines, they are having an effect already. The biggest was simply cleaning up and repairing the existing veneration systems which had been somewhat neglected.

A few amusing tales about the audit and what was found. Most of the fans in the ventilation shafts simply needed a bit of Tender Loving Care – but at Holland Park, they found that the roof vent which would feed air directly down to the platforms was almost totally sealed off, with just a tiny gap for air to enter. While no one is sure, they think the roof vent had been like that ever since World War 2, when many roof vents were sealed off. It is suspected this was to protect against a gas attack – although no one is really sure.

So, cleaning up the existing infrastructure has been underway for some time. The tube network also plans to start replacing the fans inside the ventilation shafts – starting with the Victoria Line – to new models which should roughly double the air flow in the shafts.

What about the sexy stuff?

I am sure readers will have heard of the ground water project at Victoria Station – and it has been a success. No questions – it works, and works well. The system currently draws most of its cool water from the Tyburn river, which slightly leaks into the circle line – and is drained into a sump. That cool water is used to cool systems in the station at Victoria and is very efficient in how it cools the air for minimal electrical effort in pumping water.

The system currently delivers some 60kw of cooling to the station, and plans for upgrades to support several hundred Kw of cooling will be carried out during the forthcoming modernisation works at the station.

Alas, only 3 or 4 other stations on the tube can use such a system though.

Another project is to use fairly conventional industrial air conditioners in several stations where they can mount the necessary heat exchangers on the outside. These have been installed in Oxford Circus, and will be installed at Euston and Seven Sisters. Waterloo will also get an upgrade, but not in the short term. The aim is to cool the ticket halls and also let the cool air drift down the escalators to the platform levels.

One forward thinking aspect is that the heat exchangers will be fed by a water based system so that if in the future they are able to get a reliable supply of cold water, then the roof top infrastructure can be removed. Hopefully they can find a cold water source, as conventional air conditioning units use vast quantities of electricity.

A bit Heath Robinson, but effective is to simply put fans in the stations. These do not cool the temperature at all – and indeed, the motors in the fans technically heats the stations, albeit imperceptibly. However, the fast flowing air makes people feel better and a survey found they were very popular last year. Nine stations had fans fitted last year, and we can expect more of them this year.

Another idea related to simply improving air flow inside the carriages themselves is being tested on the Central Line with a modified carriage. The Central Line trains have quite high curved windows, and on one carriage, these have been reduced in height slightly and direct ventilation slots put in its place. This has proven apparently successful and we can presumably see this rolled out slowly over time. Obviously, it only works while the trains are moving.

Ice Cubes.

OK, the one we all wanted to hear more about – because despite the sniggers there is actually a lot of very sensible thinking behind this project.

The idea is simply to have air conditioning in the carriages when trains are above ground – and to use a standard heat exchanger (or “fridge” to you and me) cool down a fluid to a frozen block – then when inside the tunnels they switch off the air conditioning and use the thawing ice block to cool air as it is passed over the heat exchanger.

The ice block stores coolth to offset the warmth of the tube network. Yes, coolth is a real word – or at least we were told it is.

Amazingly, after testing lots of exotic fluids, plain old water turned out to be the best, so the jokes about ice cubes were not that far from the truth. There is however quite a bit of work to carry out as freezing a large block and thawing it consistently is not that easy to achieve. That said, a static model is now working as a test platform, and researchers at the London South Bank University are developing a final version for the Underground.

Some mention was made of regenerative breaking which converts energy which would be lost as heat into electricity as the trains are slowed down, and Kevin was noticeably proud of how efficient the London Underground is at this. Further developments could improve that a bit more, but judging by how good they already sound like they are in this – I doubt there is a lot of gain to be had.

Another related aspect to the train machinery which lies under the carriages is a proposal for “under platform ventilation”. The proposal is simply to put ventilation under the platforms, hence the name – and when a train pulls into the station the vents blast air over the hot wheels and brakes and help to cool them down before they leave the platform.

Another idea which will be used at two stations is evaporative cooling where they pass air over a fine mist of cool water to lower the temperature. Its a sort of cheap air conditioner, but does have the subtle benefit of slightly raising the humidity level of the tube, which is currently unusually dry. Alas, London’s own ambient humidity level means this might not work that well – but they will at least try it to see what effect it has.

Quite amusingly, TfL is working with partners on the Moscow underground on this subject. Yes – Moscow. That city which is famous for being frozen during the winter is actually quite warm in summer. Normally, the Moscow system relies on the fact that the ground is so cold during winter that even the hottest summers cannot fully heat up the network before winter comes around again. However, they have had some (by their standards), warm winters recently and their own network is also starting to get a bit too warm.

Finally, we were told about an evolution of the Victoria Station trials and that is to drill very deep abstraction wells into the water table. These would then pump up cold water from deep under London, be used to cool the stations and then the warmed water would be pumped back into the water table. Tube stations being fairly long, gives them the advantage that they can extract water at one end and pump it back in at the other end, so avoiding warming up the water they seek to remove.

The problem with this approach is the lack of surface locations to use for drilling down. One idea is to try fitting a drilling rig into some of the deep level shelters which run along some of the Northern Line stations, but this is still very speculative.

If the abstraction method can be made to work, then they will need the air cooling units to be installed along the roof of the platforms. We were shown a mock-up of one installed a few months ago at the disused Aldwych station and it will be installed in the old Jubilee Line platforms at Charring Cross this year for more testing.

Kevin doubted that we will see anything public until 2012 frankly as the testing has to be very rigorous – not only to ensure the units don’t drop off the ceiling – but also to work out how to install them in the mere hours overnight when the tube is not running. Also, maybe just 10 stations would be suitable for this method of cooling.

Incidentally, the abstraction method is already in use – at the Royal Festival Hall.

So, that’s it.

Lots of repairs to existing infrastructure plus plans to improve that infrastructure over time, and then lots of innovative ideas to try out and deploy where suitable.

I think what impressed me most, was not just how they are improving the current systems – but also how open-minded they are to new ideas and it is evident that the future cooler tube will not be just one big idea – but dozens of little ideas each used where appropriate.

A Q&A session followed, and I have generally included any comments in the summary above.

A final point was raised though, and that is what to do with all the heat once it is removed from the tube network. At the moment, that heat is just dumped into London’s atmosphere – which is frankly wasteful as well as proving a planning headache in some locations.

They are talking to organizations who might like a steady supply of moderately warm air. For example, pre-heating water before its topped up to become hot water for offices and homes.

Our host from The Engineering Club finished off the evening by suggesting that we should pipe the warm air to pubs so that they don’t need those patio heaters outside. A suitable laugh, but brought sharply into focus when upon leaving the lecture I came face to face with a veritable army of patio heaters lining the pavement opposite the Building Centre – belonging to a local restaurant.

11 thoughts on “Cooling the London Underground

  1. Probably the most informative post I’ve ever read on what’s actually happening with cooling the Underground, delivered without dumbing down the reason for the gradual rise in temperature – thank you.

  2. They mentioned trying to reuse the heat for surface facilities – but it is quite difficult.

    Few buildings have communal heating supplies and also happen to be conveniently close to tube tunnels.

    Off the top of my head, the only one I can think of is the Brunswick Centre which is close to the Picadilly Line, and that would still need a lot of retrofitting to make it work again and be able to use tube-extracted heat.

    The other factor is that the houses are unlikely to want warm air during summer when the network is most desperate to extract it, so I would expect any usage would be during winter as an attempt to reduce the latent heat in the soil before summer starts to warm it up again.

  3. Fascinating. Did they mention reducing the heat from the lights in tube carriages? I’ve noticed that they get fairly warm. I wonder if using LEDs would reduce the heat? The big electronic advertising screens are also fairly war. Sure, they pay for the tube – but are they a significant cause of heating?

  4. It wasn’t mentioned as a specific issue, although there is a lot more lighting in the network today than there was in the past, so I guess it has to be a factor, albeit a small one.

    I would guess that when LED based lighting becomes affordable, they might be able to switch to that as it uses a lot less power than current lighting systems, but only if the lights can cope with the intermittent power supply.

    Hmm, I wonder if it is possible to convert ambient heat into electricity?

  5. Why couldn’t the under-platform coolers be used to provide ‘coolth’ :-) for the whole station constantly, rather than cooling the wheels, which surely would make the platforms warmer?

  6. Not sure if this blog is still visited it, but I got to thinking about this only recently.

    Isn’t the trick too cooling the underground in summer to increase the amount of ventilation in winter?

    With improved ventilation, would it not be possible to pump enough ambient cold air in to cool the tunnel and platform air to around 5-10 degrees C (a comfortable temperature in coats) for about 3 months of the year, and a further 3 months at night time?

    That would start to cool the clay which will then absorb heat in the summer months.

    1. What you have to bear in mind is that moving air itself generates heat – and in order to extract enough heat to make a difference you would have to pump in so much air at such a high volume that you would end up heating the air to the point where it loses the ability to absorb heat within a hundred yards or so of entering the tunnel.

Leave a Reply

Your email address will not be published. Required fields are marked *