Cooling the London Underground

In March 2008, I attended a really quite interesting talk on the technological challenges in cooling the London Underground network, and it has been a topic which – probably understandably – has interested me ever since. A couple of months ago, a paper was submitted to the Proceedings of ICE journal by the same people at that previous talk with a few updates.

Most of the journal paper covers issues that I have already raised in my previous blog post. After all, even with the best will in the world, a couple of years is too short a time frame to have effected that significant a change, but some improvements have taken place.

I’d suggest reading that earlier blog post, then return here to read the updates below.

—

While it is generally known that the Jubilee Line was built with an improved ventilation to help alleviate the heat build-up in the soil, I was not aware that it has “approximately the same ventilation capacity as all the other five deep lines – Central, Victoria, Piccadilly, Northern and Bakerloo – combined.” Probably goes a long way to explain the considerable temperature difference you can feel when walking through the linking tunnel between the Bakerloo and Jubilee Lines at Baker St station during the summer months!

However, if you have an older line – such as the Victoria Line – then you need to retrofit to existing structures. As mentioned in my earlier blog posting, the fans in 13 of the 17 shafts are being upgraded to boost air flow in the tunnels and evacuate the heat. The upgraded system will then be able to swap over the air in the tunnels 10-15 times per hour, which is about double the current rate. The fans are not being replaced in a row from one end to the next, but staggered around the network so that shutting down one fan doesn’t adversely affect temperatures in the surrounding tunnels. However, the new much more powerful fans also have to be fitted with sound dampers, and this is apparently one of the more complex aspects of the upgrade work.

Five of the ventilation shafts were upgraded last year in the first phase. Shafts 7–10 are scheduled to be brought into use during 2010, leaving the final four shafts to be completed by 2011.

I think the below image is instructive as it hints at the size of the fans used for ventilation – as you can see the staircase needed to get around the structure.

As mentioned in my earlier blog post, the Victoria Station was also the trial site for the very successful water cooling system that used cold groundwater that leaked into the tunnels to carry away heat. This system is still in place today and working perfectly well. The project has now been extended though – to a trial of abstracted water at Green Park station.

While the water supply at Victoria was easily available, as it leaked into the tunnels, the trial at Green Park involves drilling holes deep into the water table under London and pumping up cold water from below, using it to cool the air, and pumping the warmer water back into the ground a sensible distance away.

Boreholes running to 130 meters below ground to pull up cold water have already been drilled, and work is starting on the two boreholes needed to pump the water back into the ground, about 200 meters away from the station. They are currently finalising the design for the air ventilation units needed in the platform areas to circulate the cooled air, so it looks like the 2012 date for a live trial of a borehole based system mentioned in my earlier blog post is still on target.

They are also still looking at plans to try and reuse some of the extracted heat. It costs money to extract the heat, and if you can sell it – say to warm a nearby building – then that helps cover the costs for the tube network. Still very speculative though.

In the longer term, funding to design and implement a programme of works in support of the Piccadilly line upgrade has been provided and initial planning is underway.

The 9-page paper is available on the ICE website, with a few more diagrams than I have *cough* borrowed above.