DIY shower water heat recovery

Feb 01, 2006 31 Replies

I would avoid any version with cold pipes inside the drain pipe myself. You'd soon have these problems:

  1. soap, hair, grease build up round the cold pipe, making for much reduced heat transfer.
  2. This muck-catching arrangement causes a pipe block
  3. It also causes a foul smell outside, if you have an open drain, since much muck is perpetually trapped in there.
  4. You cant clear it by rodding.
  5. And if Murphys law holds true, your drain rod will jam solid inside the piping.

If you must use a pipe-in-pipe, have the drain as the central pipe.

I would be hesitant to make them this way as diffrential thermal expansion could cause a lot of stress, and solder is a very soft metal. Wrapping microbore round the drain pipe eliminates these problems, though it does mean a fair amount of soldering. Its possible it may work ok unsoldered, eg with some grease and zinc oxide wiped where the pipes touch, and just some solder at each end, or maybe even large a stainless jubilee clip.

BTW theres no point insulating the thing once made. If you've got insulation to use, put it somewhere on the hot pipe.

NT

I was wondering about something like this. Preliminary design was:

Hair filter.

Channeled plate to take the flow of water and distribute it into an evenly falling circular curtain of 10cm diameter (or so). Copper coil, 10cm*30cm or so of close-wound pipe.

Cold comes in at bottom of coil, and leaves top warmed by the water film flowing over it.

The plus over the american 'coil wrapped round pipe' should be better heat transfer, without a long vertical drop. The minus is of course need for occasional cleaning.

To address the cleaning issue, I was planning either to make the coil completely removable, with something like (but not) hozelok connectors, or to have the channeled plate removable, so that you have access to the inside and the outside of the coil in the 20cm*30cm well it sits in.

(The well is designed to never have water sitting in it)

I guess it would work, yes. Its adding maintenance though. But a 1' drop, if your bath's on the ground floor you probably wont have that much room.

Ah: because it would fil with muck and gunge, and people being busy with more pressing things, you'd end up with such units getting stinky. I would want to keep any possible stink gen on the other side of the water trap. But that then makes it inaccessible...

NT

A water trap could easily be done simply by making the inlet do:

| | \wwwwwwwww|wwwww|wnwww/ .\___________________/. : : : : O O : : O O

(O = pipe, : = falling water, www = water surface, n = water spirit (nymph, or similar)

That's what I'd be interested in seeing. What actual potential heat gain is there? How does that translate into energy saved (if you're into being green) or money saved (if that does it for you)

the one time I calculated this, the payback was excellent. These things do look like a genuinely good idea.

NT

I'm a little unclear about your diagram, so I might have misinterpreted it, but its starting to look complex to construct and hard to get to the muck storage to clean it. It seems theres no free lunch.

NT

Hokay. Bigger diagram.

Shower floor. Removable plug

_________________ ____ _____ ___________________ || |H| || |- | | -- | \_____/ | | : : | | | warm out>| O O | | : : | | O O | | : : | \ O O /

ahhhh, comprendi.

May I suggest a tweak: something closer to 6" high and nice and wide, so you get plenty of coil in there, and lots of surface area. And so the bath isnt half way up the wall :)

NT

I'm not sure. At some point, as you increase in diameter for a given flow, the curtain of falling water is going to be harder and harder to get even, perhaps even leaving bits of pipe 'dry'. Then there is the issue of as you decrease the number of pipes a given bit of water flows past, you increase the temperature drop from one pipe to the next bit down, meaning that heat transfer has to be much more rapid. I _think_ that for a given flow of water, thin and long is better than short and wide. I've got to do some simple simulations though.

true. I expect in practice it would snake about, but theres no g'tee.

erm... with wider dia, less volume of water flows per second over each degree of the pipe pancake. It also flows less distance from top to bottom. I think it balances out ok.

maybe, but I dont think this is the decider. Having the bath on the floor is more important imho than a slight thermal difference.

NT

I'm unsure - anyway, where are the breakpoints? How much more efficient is 4 turns than 2, ...

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required