HW cylinder drain

[Fred]

Hi,

The inlet pipe to my HW cylinder runs behind it, right at the back of the airing cupboard where it is totally inaccessible.

I'm decorating a plasterboard wall in the bedroom and I wondered whether I should take the opportunity to cut a small hole through the wall and fit a 22-22-15 tee, run a 15mm pipe from the branch past the side of the cylinder to a drain c*ck at the front of the cupboard?

Now I'm not expecting to need to drain the cylinder but if I don't fit a drain, you know something will happen!

If I do fit the tee, part of me wants to use an end feed fitting so that it is securely in place. I worry that if I use a compression fitting and the pipe gets knocked, the fitting might loosen and weep and it will be behind the cylinder and completely inaccessible.

OTOH I don't know whether I could solder onto the pipe because one option to stop a flood would be to freeze the pipe before working on it and the ice plug won't last long if there's a blow torch about! But if I drained the cylinder some other way (how?), the pipe would be wet and wet pipes don't like solder.

I have also heard that solder will not "stick" to old pipes. I'm not sure that's been my experience, and I always use wire wool and flux first, but again, I don't want a leak where I can't see or fix it.

One advantage of fitting a tee is that one day I will get round to fitting a pump and a return pipe to take hot water to and from the kitchen tap. At the moment it takes forever for hot water to get there. This would give me somewhere to take the return flow to.

What do you think?

TIA

[grimly4]

That's what I've done with mine, with remote drain via full-bore lever valve. I've wrestled and strained with too many of the crappy little square-topped brass valves to be arsed with one on my own installation.

[Tim Downie]

If it was just for drain-down, it might be easier just to use compression fittings at the HW outlet on the top of the tank. Then, if you do need to drain, just undo the fittings and drop a hosepipe in and syphon the tank. After all, it's not something you need to do that often.

Tim

[Dave Liquorice]

One assumes the circulation pipes will be well insulated to minimise losses and thus running costs. Most of the time the water coming back will be not much cooler than that going.

Wonders if an instaneous electric heater in the remote location with a control system that only energises it when the supply is below a given temperature would be cheaper to run? Maybe something with a thermosatic mixer valve. haven't thought that through though.

[fred]

If your water is returning cold then you won't be maintaining hot water at a decent temperature at the delivery point. As Dave points out, you want the circulation piping to be super-insulated to avoid excessive heat loss.

Low heat loss in turn that means you don't need much flow so it can be a tiny return pipe and pump.

You can return water near the top using a modified surrey or dursey flange to drop a narrow pipe down a third into the cylinder. This can be created using a reducing T mounted vertically above the outlet with hot water coming off the branch and a length of 8mm microbore coming down through the through part of the T to carry the return leg water. The microbore is terminated in the cylinder with some compact diverter arrangement (see surrey or dursey flange pics) to stop return water force and flow destratifying the cyl.

That said, hot water circulation is a rich man's game and best avoided ;-)

If you can, and have the water head to do it, swap the run to the kitchen to 15mm which will have less than half the cold water in it that a run of 22mm has.

[fred]

In article , fred writes

Dursey flange should read danzey flange.

[Andrew Gabriel]

Full bore valves are not always good when virtually never operated, as they tend to sieze up with non-use.

[Fred]

Yes. It will be.

I suppose I would need to connect all the plumbing and do tests on the return temperature. If the water was 60C flow and 50C return, would it be advantageous to return the return to the cylinder to reheat it as I understand you should heat the water to 60C to kill bacteria. Or perhaps I worry too much!?

You can get under sink electric heaters. I guess the advantage is you only heat what you use but I thought that the elements do not last very long in these? Also I imagine that to get a good flow you need a fairly powerful element, so you probably need to run a cable specially for it.

[Fred]

You are quite right that before doing anything else, I intend to replumb the pipes to the kitchen. At the moment there is an (original) section that runs through the concrete floor of the kitchen, which I am sure is responsible for much of the heat loss. Redone either above floor level or in the ceiling overhead and lagged, I hope this will improve things dramatically.

I could use microbore for the return but can you get lagging that size?

TIA

[fred]

Sounds good. If you're really stuck then you can use a 22mm run as a conduit for 15mm plastic pipe[1] but I wouldn't use such a run on circulation the heat loss would be too great for my bank balance.

[1] Do-able but not easy.

The alternative is to run the feed and return together and insulate them in one cover using the next size up in pipe insulation, say 28 for 22+8 or 22 for 15+8. Just make sure the seam is closed tight (cable ties every 6" are good and tape the seams/joints with a v good quality duct tape for longevity. Go for the thickest insulation you can, I have 25mm wall stuff on my 15mm runs, only a few v short runs of 22mm which have

19mm wall stuff.

Don't worry too much about the 'cold' return being bunched with the hot or the possible voids inside the insulation, neither should be a bigee.

[Fred]

I have been using google to find out more about flanges. There seem to be (amongst others) York and Surrey types. Is the only difference between these two that one has male threads and one has female threads or is there more to it than that?

Why are there so many types? How do Essex and Durzey differ from York and Surrey flanges?

The applications I have found seem to use the flanges to draw off water rather than return it, but I can see how it would work in reverse. The pages I have found show the flange being used to supply shower pumps. I am puzzled by this; doesn't the narrow internal pipe cause some restriction to flow?

Back to my scenario, I'm still not entirely clear what the problem would be if the water entered the inlet at the bottom. Hot water rises, so wouldn't it reach the right level? Is the concern that if the bottom of the cylinder was cold, the hot water would be cooled as it entered? Is the advantage of using the flange that hot water returns to hot water as is not cooled?

If I do ever get round to doing this, I will do it properly with a flange but I'd just like to get the theory right.

TIA

[Fred]

Is the same true of the reduced bore isolating valves? I would think the mechanism would seize in either size?

TIA

[fred]

Essex flange penetrates the cylinder but the others make access via the top outlet to avoid cutting into the cylinder.

The beauty of water pipes/flanges is that they are bidirectional ;-)

The idea of the short pipe is to grab water that is free from air, just below the top. Although the pipe is narrow, it is short so the restriction is less and these are normally used for pumped showers so the small restriction is easily overcome by the pump.

Also, see the pic of the surrey flange on this page where it narrow through the open flange but opens out again once past it.

that I didn't suggest using any of these flanges for your job but suggested copying the design to get a small pipe into the cylinder and place the outlet lower in the cylinder to suit the return temp.

You do not want hot and cold water to be mixed up when there is a high demand for hot water. The ideal is for hot water to remain stratified at the top and for cold to gently replace it at the bottom, then you have a whole tank off hot water that then suddenly goes cold when the hot water 'runs out'. If you create any kind of stir in the water to mix hot and cold then you end up with a short run of hot water then receive progressively cooler water as you draw more and more off. Returning hot water at the bottom will stir the cylinder up, causing undesirable mixing.

See the stratification section of this article for a bit of background:

[Andrew Gabriel]

Not in my experience, but that's not particularly extensive.

Full bore has bigger balls, which have more surface area to seize up. At least, that cause fits with my observations, even it it's the wrong cause.

[Fred]

True, but is there a reason I could not use an off-the-shelf one?

I shalln't return the water at the bottom for the reasons you have told me but I wonder whether flow from bottom to top can ever be useful whilst heating? Would a slight current help eliminate hot spots on the immersion heater or coil from the boiler?

TIA

[Fred]

An update: there is a drain c*ck behind the cylinder but it is so low it almost touches the floor. I realise it needs to be low to achieve a complete drain but it is so close to the floor it is difficult to fit a spanner underneath to turn the square to open the drain (the square end is pointing down towards the floor).

It doesn't help that the cylinder on one side and the wall on the other do not give much room to turn a spanner. I may buy a cheap spanner (8 mm?) and cut it short with an angle grinder so that I can turn it with less obstruction.

I unscrewed the nut on the top of the cylinder, which appears to be a standard nut off a 22mm fitting. Looking into the cylinder I could see a hard green deposit partly blocking the outlet. Unfortunately this deposit is reducing the bore of the outlet just enough to stop me pushing a hose pipe through!

If I can't get the drain c*ck to turn, I think I will have to buy some narrower plastic tubing and siphon it very slowly.

I'm wondering whether the scale is a problem. I don't really want the expense of a new cylinder just yet. Is there any way of descaling it? Google has found "fernox ds3" but it says not to use on enamel, which means I can not just open the taps and drain it into the sink or the bath (I would have to use a bucket). After use it requires a lot of flushing to completely get rid of it. Is there a way that is less hassle? Citric acid perhaps, but I suppose I would need bucket loads.

TIA

[Andrew Gabriel]

It will damage vitreous enamel (discussed recently in another thread), but it won't harm a porcelain basin or stainless steel sink. It contains an indicator dye to show when it's used up (green OK, blue used-up). If your basin has a thin coating of hard water scale (they often do), it will stain it blue, but you can wash that off by using DS3 (or any acid or scum cleaner) to dissolve the scale off.

Anything you use will need to be flushed out. DS3 contains both an indicator dye and an added scent so you know when it's flushed. It's not particularly toxic (many people have accidentally drunk it as a result of forgetting the kettle is being descaled when they make a hot drink, without any subsequent ill effects).

The other risk descaling an old cyclinder is that it might have pinhole leaks which have self-sealed with scale, which will be unplugged by descaling them. If you do descale it, I would suggest filling it only enough to cover the coil - that's the only bit worth descaling (and the immersion heater perhaps, but not worth filling the cylinder to the top just for that).

[Fred]

Thanks. Do you remember what the thread was, I'll have a look for it. It's good to know it is only certain types of enamel that are a problem.

Damned if you do and damned if you don't, it seems.

Thanks.

[fred]

You could certainly start that way but I would probably extend the pipe in the cylinder so it was lower, so that if the return temperature was cooler, it would be returned lower down and not be immediately sent back out on the circulating loop to return cooler still. Note again that the return pipe should not just have an open end, it should have a bend or a baffle to stop the returning mid temp water being jetted to the bottom of the cyl and again upsetting stratification.

Dunno but I think it already happens through convection, coil/immerser heats water in immediate contact, this water then rises to be replaced by cooler and so on.