Hot water system problem...

[Andrew L]

Dear all,

I need some help diagnosing a problem I have with my hot water system.

I've got a cylinder tank in the airing cupboard which has above it two square tanks, one large one small. I believe the larger is called the feeder tank. As hot water is drained from the cylinder tank water is drained from the feeder tank into they cylinder, which in turn is filled up from the cold water mains supply up until a certain level (about a

1.5 inches below the overflow) at which point the float valve closes.

This all seems to work as described. If I don't use any hot water for about 4 hours though the water level in the feeder tank has reached the overflow and starts trickling out. It's a tiny, tiny trickle, so much so that the water doesn't pour out of the end of overflow pipe but trickles out the end, back along the bottom and is absorbed by the wall. The plaster on that wall is blown. :-(

At first I thought that the float valve was shot, so I put in a replacement, but that hasn't changed anything.

Any ideas as to what the problem might be, or what I should try next?

Many thanks

Andrew

[Dave Baker]

The overflow should vent outside and there should be enough downslope on it so that water can't run back down the pipe. As to why the level rises there are a couple of possibilities.

1) Hot water expands so you tend to find the level in the header tank rises a bit anyway as the cylinder heats up. Set the float position lower. 3 or 4 inches below the overflow is about right. 1.5 is too close. If the float arm is plastic there should be an adjusting screw on it and you just bend the brass ones by hand. 2) If there is a steam vent pipe from the cylinder to the header tank does any water trickle from that? A pump on too high a setting or a badly installed system can lead to this.

[Tom]

but trickles

Can I suggest as a temporary measure until the problem is cured, that you tie a little bit of thin string or strong cotton around the outlet of your overflow pipe so that the trickle is drawn down the string and clear of your wall. Regards Tom

[Andrew L]

Sorry, I probably wasn't clear on that point. The water leaves the feeder tank via the overflow. The volume of water is so small that when it gets to the end of the overflow pipe outside the house it goes down the cross section of the pipe, back along the underside of the outside of the pipe until it gets absorbed into the wall. I guess any physicists here will be able to explain it in terms of surface tension?

I'll set it to around 3/4 inches below the overflow.

I haven't seen any water coming from the vent pipe, but I will keep an eye on it just in case. I'll let you know what I find out.

Thanks for your help,

Andrew

[Andrew L]

What a neat idea! Thanks.

Andrew

[Owain]

"Tom" wrote | "Andrew L" wrote | but trickles | > out the end, back along the bottom and is absorbed by the wall. The | > plaster on that wall is blown. :-( | Can I suggest as a temporary measure until the problem is cured, | that you tie a little bit of thin string or strong cotton around the | outlet of your overflow pipe so that the trickle is drawn down the | string and clear of your wall.

Or, as it's probably a plastic pipe, form a little lip at the base of the pipe using pliers and the heat from eg a cigarette lighter. Or push an elbow fitting on to the end.

Owain

[John]

If you have replaced the float valve AND there is no sign of it dripping when up to level then you may have a leak in the coil inside the cylinder which is allowing water from the heating system to make its way through into the hot water (tap) side. This may be evident from a dribble from the heating header tanks ball float valve without attendant rising level in that tank. If this is the case then you will have to replace the cylinder. Continued leakage will cause damage to the inside of your boiler and rads etc and contaminate your hot tap water

[Andy Wade]

But only if the CH F&E cistern is at a higher level then the DHW cistern, surely. If both cisterns are standing on the same platform that the flow would go the other way, would it not?

[Andrew L]

John - thanks. Tonight I'll stop the cold water feed to the feeder tank just to double double confirm that it's not coming from the float valve (I'm 99% sure it's not, but could be that the replacement, though new, is faulty)

Andrew.

[John]

Thats why I said "may" have a leak

[Colin Wilson]

This might be a long shot, but hey, i`ll give it a go :-}

Do you have any hot taps or pipework above the level of the tank ?

I`m thinking if the water might be siphoning back to the tank for some reason...

[Andrew L]

Thanks for all of the replys to the original post. Having done some experimenting I've done/found the following out:

- the level at which the float valve cuts off is about 3 inches below the overflow pipe.

- the feeder tank fills up irrespective of whether the cold water feed is turned on or off. ie Overnight wednesday it filled up with the stopcock for the cold water feed turned on. Last night it did the same with the cold water feed turned off.

- the water in the feeder tank is lukewarm. Not boiling hot, but slightly warm to the touch.

- there's no water coming from the steam vent pipe.

- there are no hot taps or pipes above the level of the tank.

Does anyone have any further ideas as to what may be happening/what I can try before I call out a plumber?

Cheers,

Andrew

[Set Square]

Is the water level in the small tank higher than that in the large one? [The small tank is the fill & expansion tank for the heating/hot water primary circuit]

The water in the primary circuit is heated by the boiler, and is pumped through a coil inside your hot water cylinder - enabling it to heat the domestic hot water without mixing with it. However, if that coil develops a leak, the waters *do* mix - and water flows from one circuit to the other. The direction of flow is determined by the relative pressures in the circuits. Whichever circuit has a higher water level in its header tank is at a slightly higher pressure than the other one - and water flows from high to low pressure.

If your cold header tank continues to fill with the water supply turned off then - discounting a small rise due to expansion - the water *can* only be coming from the other circuit. The fact that the water in the cold header is lukewarm re-inforces the notion that it is coming from the primary circuit.

You say that when it reaches overflow level, the drip rate is very slow. Could it be that the overflow level in the big tank is only *just* below the static level in the small tank. If so, the pressures would be very nearly equal by the time it rose to this level.

If this *is* your problem, you'll need a new hot water cylinder!

[John]

One other possibility comes to mind - is there a leak between a cold mains fed input and the hot water service via a defective mixer tap or temperature control mixing valve located anywhere in the house? Try turning off the cold water main at the stop tap and observe the header or feeder tank as you call it.

[Andrew L]

Thanks Set Square,

No the water level (and overflow) of the small tank is below the water level of the larger tank.

Cheers,

Andrew

[Set Square]

And it's the larger tank which continues to fill, and overflows - even with the water supply turned off?

Then I'm flummoxed!

The suggestion made by someone regarding some sort of cross-coupling between the mains cold water system and low pressure hot water system seems worthy of investigation.

[Andrew L]

Downstairs we have a gravity fed mixer shower. Water comes direct from the hot water tank into one side, and from the mains feed into another. The temperature control is separate from the on-off control. Normally we keep the temparature near maximum hot. Overnight last night as a test I set it to cold...and this morning the level in the feeder tank hadn't risen a millimetre.

This looks like it could be my chief suspect although quite exactly what is happening I'm not sure.

Thanks,

Andrew

[Andrew L]

Downstairs we have a gravity fed mixer shower. Water comes direct from the hot water tank into one side, and from the mains feed into another. The temperature control is separate from the on-off control. Normally we keep the temparature near maximum hot. Overnight last night as a test I set it to cold...and this morning the level in the feeder tank hadn't risen a millimetre.

This looks like it could be my chief suspect although quite exactly what is happening I'm not sure.

Thanks,

Andrew

[Set Square]

It does indeed! It sounds as if somehow, even with the shower turned off (with nothing coming out of the shower-head), high pressure cold water can flow across to the hot water side - and push hot water back up into the header tank.

I guess that the mixer valve wasn't designed to cope with unequal hot and cold pressures - and does the mixing *upstream* of the on/off valve.

[Andrew L]

The hot and cold feeds go into a mixer tap (the temperature control) and then to the shower head via an on/off tap. When the shower is switched off it's almost impossible to turn the temperature control from hot towards control - it's as if you're fighting a great pressure.

It's been operational for nearly two years and not caused a problem before. However the water is exceptionally hard around here (it's destroyed a dishwasher in four years) so I guess that there may be some corrosion in there somewhere. Keeping the temperature tap on cold is at least a temporary fix.

Many thanks to all who offered advice.

Cheers,

Andrew