Central Heating Expansion Tank Keeps filling Itself!

[BraileTrail]

OK, I have an old (25 years plus) semi-gravity gas fired central heating system. In the loft are the usual cold water storage tank and heating system expansion tank.

The overflows for the 2 tanks are combined in to one outlet that comes out over next doors garage! Not a great idea and I will be changing that soon.

Next door told me the overflow was running so I had a look in the loft and indeed the expansion tank level was mm below the overflow pipe. The valve was very old, scaled up and had no service valve. I fitted a service valve and changed the valve for a modern diaphragm type and took out about 8 litres of water while the system was cold.

Checked next day and the level was again mm below the overflow! Turned off the service valve and took out another 8 litres of water. Next day same again, the level is mm below the overflow.

I started to think that the old ball valve in the cold-water storage tank was not working and that it was overflowing and, via the linked overflows, filling the expansion tank. Looking at the fall on the overflow pipe this is very unlikely, but couldn't think of anything else. Today I changed the ball valve on the cold-water storage tank and set the level about 75mm below the overflow, everything seems fine there. Definitely not overflowing.

So, once again I drained 8 litres out of the heating expansion tank while the system was cold and left it with the service valve closed. Over a 5 hour period the level has risen by 30 mm with the heating and water off and the service valve closed! The cold water storage tank level is still 75mm below the overflow.

My only other thought is that the coil in the hot water tank has corroded / burst and water is getting in to the system that way. Any other ideas or thoughts on the matter would be appreciated.

BraileTrail

[Tim Downie]

Could it be that it's the valve on your HW header tank that's leaking and back filling the CH expansion tank through the siamesed outlets?

Tim

[BraileTrail]

Tim,

In message , Tim Downie writes

It could be, please read on. ;)

BraileTrail

[David WE Roberts]

What are the relative levels of water in your hot water and central heating expansion tanks?

If the heat exchanger coil in the hot water tank is leaking (which is not uncommon in systems of that age) then as far as I can see this would over time level the water in the two tanks as they have become directly connected.

Presumably if it is a very small leak then the central heating system would suck water in when cold (level in header tank goes down then comes up again to match hot water header tank) then when heated rapidly by the boiler it would expand and fill the small central heating header tank.

As you say, there doesn't seem to be any other way for water to get into the central heating side of the system.

[BraileTrail]

Thanks David.

In message , David WE Roberts writes

The expansion tank is sitting on the same plinth as the cold-water storage tank so their bases are at the same level. The expansion overflow level is below the full level and the overflow level of the cold-water storage tank.

The overflow from the cold-water storage tank drops down vertically to where it tee's in to the overflow from the expansion tank.

That is my suspicion, of course it can't level it because the expansion overflow pipe drains it off.

One thing I don't understand is that the expansion tank is just that, to cope with the expansion when the water in the heating system gets hot, I'm surprised (but pleased!!) that the water doesn't expand and come over the top of the tank before the overflow can empty enough of it.

Again, that follows my reasoning. Sounds like a new tank may be in order.

Can anyone think of anyway of testing the hypothesis before changing the tank? So far there is no evidence of any leakage the other way, no brown staining in the hot water or anything.

BraileTrail

[Peter Parry]

Turn off the water to the main tank and drain enough from a hot water tap to bring the water level in the main tank below the level of that in the expansion tank

Does the water level in the expansion tank now fall so both are more or less level?

[ransley]

What feeds new water the boiler, is it an auto feed, they go bad. Shut off the feeds. How many stories to the house, what is boiler water level, I go up 35+ft but only need about 15lb of pressure in the boiler when hot. Does the pressure relief on the boiler ever go off, is there any indication on the boiler it does release, as in deposits around the valve, lowering pressure may be the issue, you only need enough pressure to get water to the highest radiator.

[BraileTrail]

In message , ransley writes

The expansion tank! It is a normal 2 story house.

It's an open-vented system, it's all at atmospheric pressure.

BraileTrail

[Bob Minchin]

Tie up the expansion tank ball valve and ascertain that the water supply to the tank is truly cut off. Siphon water out of the expansion tank, or draw some off at the boiler drain c*ck which ever is easiest.

If after some time the expansion tank fills up then you can be certain you have a leak between primary and secondary circuits.

Quite a normal fault and unless you are very lucky, the new tank will have inlets and outlets in different places so prepare for a fairly significant job grovelling in your airing cupboard replumbing. Fit a new long life immersion heater whilst you are at it as it will almost certainly be corroded too. A replacement tank fitted with sacrificial aluminium anodes will last longer if you are planning to stay in the house 10yrs or more. Good luck

Bob

[The Wanderer]

If the service valve for the CH header is closed and the level is still rising, there's only one place of common coupling between the two systems, and that's the coil in the cylinder. Had exactly the same thing a year or two back. My system is about 35 years old.

[BraileTrail]

In message , Peter Parry writes

Nice one Peter! My job for tomorrow morning. If I turn it off before we shower in the morning then we wont even waste the water!

BraileTrail

[The Wanderer]

Not necessary to prove where the problem lies. If the service valve for the expansion tank is off and the level is still going up, there's only one point of common coupling between the heating and domestic hot water circuits, and that's the coil in the cylinder. New cylinder it is! :-(

[BraileTrail]

Thanks, Bob.

In message , Bob Minchin writes

That much I know to be true already. With the service valve off the level in the expansion tank rose 30mm in about 5 hours. The there is clearly a leak some where and the most likely suspect is the hot tank coils.

Range will make a hot water cylinder with custom tapping's for you if you ask. It doesn't cost anymore, but they are expensive cylinders already.

BraileTrail

[BraileTrail]

In message , The Wanderer writes

I hear what you and Bob say, but I would just like to "see" it before launching in to a new cylinder.

BraileTrail

[Peter Andrews]

Be aware that, at the moment, the flow is only from the hot water to the central heating - if you change the water levels, such that the cold tank level is lower than the expansion tank, you will CONTAMINATE your hot water system with any chemical inhibitor that is in your CH system.

Peter

[BraileTrail]

In message , BraileTrail writes

Of course they can't do much about the heat exchanger tapping's I suppose.

BraileTrail

[BraileTrail]

In message , Peter Andrews writes

Thanks for the warning, but I doubt that the flow is currently one way only. Osmosis is going to ensure some mixing of the waters I would think.

By lucky chance the system currently has no inhibitor I drained the system to do some work several weeks ago and ran the system for a while without just to check for leaks. I decided that I wanted to convert the system to C-Plan to get independent control of the hot water and never did put the inhibitor back in.

According to my neighbour the overflow leaked when the previous owners were here and he couldn't get it sorted. So neither the work I did or running the system without inhibitor for a short time has caused the problem.

BraileTrail

[geoff]

In message , BraileTrail writes

An interesting concept

total bollocks, but interesting

[BraileTrail]

In message , geoff writes

It is, fascinating.

Rather harsh! Anything else useful to add?

How about diffusion then?

BraileTrail

[Dave Liquorice]

Well not quite, the pressure in the system at the bottom has the head of water to the level in the expansion tank as well. Reducing to atmospheric pressure at the level of the water in the tank.

This is why the flow with a holed coil in a cylinder is generally only one way from the HW system into the primary. The cold water storeage tank is normally much larger than the heating header tank thus the level of water in the CW tank is higher than that in the heating header tank. Thus the pressure in the coil is lower than that in the HW tank, may only be a foot head of water but it's enough to ensure the unidirectional flow.