Wednesday, April 25, 2012

8 April 2011 – Egout idea


As something of an aficionado of alternative energy sources, I was mesmerized by a Reuters article earlier this week that reported that various public buildings in Paris - schools, swimming pools, and even the Presidential Palace - are soon going to be heated by energy drawn from the city’s legendary sewers.

No, wait - don’t click out!  I’m not kidding.  Here’s the article:

Heating buildings with energy drawn from sewers - 'cloacothermal' power, if I may be permitted a neologism - is perfectly feasible, and actually quite elegant.  The Paris sewers channel a massive flow of waste - 285,000,000 cubic metres per year, according to the article - which remains at a steady temperature of between 12 and 20 degrees Celsius.  Since one calorie is the amount of energy necessary to raise the temperature of one gram of water by one degree Celsius, and a cubic metre of water (we’ll forget the icky solid stuff for now) masses a metric tonne, that means that the sewers transport a minimum of 342 trillion calories, or 1.4 quadrillion joules, of energy every year.  That’s about a third of a megatonne in nuclear weapons terms; or, in electrical terms, a little less than the annual electrical consumption of Togo.  At present, all of that energy goes to waste, and it’s a great idea to try to capture some of it. 

This is how it works.  A metal plate containing hundreds of internal channels - a heat exchanger, more or less like the radiator on your car, only a lot bigger - is submerged in the waste flow.  Pumps channel a heat-exchange fluid (usually a water/ethylene glycol mixture or some similar antifreeze solution) through the plate, forcing cold liquid down into it where it is warmed by the waste flow, and bringing the warmed fluid back up into a heat pump.  The heat pump works like a refrigerator or air conditioner; the warmed heat exchange fluid is channelled into an evaporator where it heats refrigerant in a copper coil.  That refrigerant is then forced through a reversing valve into a condenser, where the gas is compressed, forcing it to release the concentrated heat.  That heat can be used to warm just about anything.  For a building, it is used either via forced-air circulation, or - since this is Paris - probably an hydronic system where the heat is circulated to conventional radiators via another water or water/antifreeze loop.  As the engineers cited in the article rightly point out, there can be no odour, since the original heat source and the destination are totally isolated.  With decent insulation, the heat can actually be transported a reasonable distance with minimal losses.

This is precisely how a lake-plate heat pump works, except that for a lake system, the heat exchanger plate is submerged in a large body of water, while in the proposed Paris system, the plate is submerged in…well, not to put too fine a point on it, but yesterday’s boeuf bourgignon.  The best part is that there’s no well drilling involved; you just drop the plate in the waste flow, connect the pipes, and you’re off to the races. 

For those who’re interested, I’ve included a link to a brief bit of literature on plate heat exchangers and how they work. (Note A)

This is all very cool.  I foresee only three possible problems.  The first is mentioned in the article, which notes that the sewers have occasionally been home to “rats, pickpockets, intrepid tour groups and the odd corpse” (not to mention werewolves and tragically disfigured composers).  None of these are likely to be as much of a problem as vandals.  Lake plate systems and their piping are tough, but they’re not invulnerable, and in the sewers they will be exposed to mischief-makers.  This is not an insignificant consideration in a city that sees more than a thousand cars subjected to multicultural incendiarism every year.

The second problem is clogging.  In order to maximize thermal transfer, a heat exchanger - like your radiator - circulates fluid through pipes surrounded by thin vanes designed to maximize surface area.  This means that the plate contains many thousands of tiny channels that are easily clogged by debris.  There’s a lot more debris in a sewer than at the bottom of a lake.  In order to maintain system performance, the plates will have to be cleaned regularly.  Depending on how bad the clogging gets, it might be easier to simply swap them out for new ones and take the old ones away for a thorough cleansing and refurbishment.  The cost of programmed maintenance has to be figured into the cost-benefit analysis for the system (and maintenance costs are always underestimated).  And I don’t know about you, but cleaning or swapping out heat exchanger plates that’ve been submerged in a sewer is not a job I’d be lining up for.  As one geothermal experted I know put it, "I wouldn't want to be the guy who has to de-scale the plates."

The third problem is thermal depletion.  The size of the plate determines how much heat you can transfer - but for cost reasons, plate systems tend to be sized as small as the designer can get away with.  On very cold days, the demand on the heating system will be high.  With a very efficient heat pump operating on a water-glycol loop, it is possible to draw so much heat out of the circulating fluid that its temperature drops below zero.  When this happens, ice begins to form on the submerged plate, causing a collapse of system performance, and also causing the plate to begin floating.  If there are enough plates sunk in the sewers, it might even be possible to draw so much heat out of the effluent flow that the sewers begin to freeze up - and when sewers freeze up, they back up, with unpleasant results for upstream customers.  Fortunately, there’s an easy fix; if the heat pump system is designed for air conditioning as well as heating (and given how more than 14,000 people died in Paris in the 2003 heat wave, a little air conditioning might not be a bad idea), you’d just have to reverse the system and run it in a/c mode for a few hours, sucking heat out of the buildings and forcing it down into the plates to melt the ice and get the system running again.  Doing so would make the Elysée a little chilly for Sarko and the missus, to say nothing of the swimmers in the solidifying public pools or the shivering tots in the sewer-heated schools - but that’s what happens when engineers have to fix problems created by mathematicians and physicists.  “Elegant, fast, functional - pick any two.”

One other thing about thermal depletion: the “material solids” contained in the effluent flow in sewers are broken down by bacterial action, and bacteria do their best work under defined conditions of salinity, pH balance, and of course temperature.  I’m just spit-balling here, but there’s a reason you don’t sink lake plates in the settling ponds at sewage treatment plants.  Chilling the Paris sewer water down by a few degrees could impede bacterial action, with unpleasant results.  Time will tell, though, I suppose.  I’m certain the planners have thought of this.

Anyway, I say go for it.  It’s brilliant.  The Paris sewers are more than a monument to the eternal glory of the Ville Lumière; they’re a precious natural resource, and there’s no reason to lose all of that glorious, bio-generated and nicely-packaged heat into the general environment.  I suppose it won’t be long before restaurants start displaying signs saying, “Please try the curry - you’re heating the Presidential Palace!”