Draining central heating

[Adrian]

Hi All

Looks as though my frinedly local plumber has pushed off and isn't answering my texts, phone messages or emails.....

I've been waiting for him to come round and flush the heat pump / central heating system he installed - but looks as if I might be waiting some (further) time....

So - I'll do it myself ! I've worked on central heating systems before - so I'm assuming that the 'secondary' side of this system will be much the same as other gravity-fed systems that I've seen.

We have fairly acidic water here (well water) - but, since the system was installed, we've added a dosing system that knocks the pH back down towards neutral or thereabouts. Trouble is, the ch system was filled with the acidic water and left to sit for a year or so before we bought the house, and I know that there's an amount of black crud in the system. We've been here 18 months - system was installed maybe

12 - 18 months before that - was originally oil-powered.

To make it more interesting, the system's been pumping over into the header tank (lots of lovely oxygen!), and the header tank's growing brown furry slime....

So - as it look like it's down to me to fix it....... here's the plan

- comments appreciated.

1) Stop the pumping over - reducing the speed setting on the two circulation pumps for starters and seeing if that does the trick.

2) Having achieved '2' - drain the system (having turned off the heat pump !). At the moment there is no drain c*ck on the system - so I was going to add one of those 'self-cutting' washing machine taps on a handy 1/2" ground floor radiator tail and take the spare end out through the wall for ease of draining.

3) Refill the system with fresh water, run for a couple of days. Try to force the water through some of the more distant radiators which may be more sludged than others...

4) Drain again. Repeat 3 & 4 until water runs clear

5) Fill, dose with Fernox, and leave well alone

Any comments, recommendations etc ?

Thanks in advance Adrian, Ireland

[AlanC]

Is the brown slime a sign of pumping over? I found a some in my header tank at the weekend. It was like the skin you get on custard when you leave it to cool (I suspect it didnt taste as good though!). I scooped it all out and the water that was in the header tank (it was quite black) and refilled. There is plenty of inhibitor in the system as I put 2 lots of Fernox in it a couple of years ago when I half drained it down (just to be on the safe side!!)

[Adrian]

Well - that was just my guess......

Odd stuff - felt kind of 'soapy' to the touch - and solid enough that it biult up quite a thickness on the float....

Adrian

[David Hansen]

On Wed, 05 Mar 2008 12:07:18 +0000 someone who may be Adrian wrote this:-

Before 2) look at the pipework and work out why it is pumping over. Probably this is because the feed & expansion and vent pipes are connected in the wrong place. Generally they should be on the suction side of the pump; so that the system is at the highest pressure on discharge from the pump, the radiators and so on are under pressure and the pressure in the system gets down to just the static pressure as it gets to the suction side of the pump.

[TheScullster]

"Adrian" wrote

The slime is likely to be fungal growth. When you've sorted the other problems, at final fill, add anti-bacterial agent (talk to Fernox they're good at that stuff). If there's any amount of black sludge, take off all rads and flush through manually. Run water out of each rad point until clear. Re-assemble system then chemical clean, flush etc.

Phil

[Adrian]

Ah - thanks !

OK....

I'm hoping that there won't be too much of that... but we'll see what falls out when I drain it....

Sorry - not quite with you there

Yes - I asked at our local friendly plumbers' merchant this morning - enquiring after some sort of 'flushing/cleaning additive' - and got blank looks. They had some 'anti-scale' additive - but we don't have hard water - just acidic (though a system refill now would be done with our treated water which is roundabout pH7)

Thanks Adrian

[Roger Mills]

You certainly need to stop the system from pumping over. If there's a design fault, you may need to re-arrange the pipework to achieve that - slowing the pump down may not be sufficient.

I wouldn't leave a self-cutting washing machine tap in circuit - it may leak where it clamps on, and the bit of pipe which it punches out may cause a flow restriction. By all means use one to drain the system - but then, once you have drained it, replace that section of pipe with a bit containing a proper drain fitting - or just a tee with a branch leading outside, with a tap somewhere in it.

The best way of cleaning out the radiators is actually to removed them (easy with the system drained), take them outside and flush them through with a hosepipe. This will be far more effective than trying to force water through them in situ.

[Adrian]

Understood - but if it can be sorted by pump speed alone then that'll be great . This was the missing plumber's first suggestion - just before he went AWOL....

The system has twin pumps - one for CH and the other for DHW, both of them were set to maximum - so I've slowed them down to the mid-speed setting. Baled out the header tank this afternoon - we'll see if the change in pump speed stops the header tank from filling to the overflow point... can but hope !

Had further thoughts about that one.... The heat pump is actually in my Glass Studio (timber outbuilding) so all the system pipework is available there. I think that might be a better place to put the drain c*ck - after all - it's 'outside' to start with - and saves drilling a hole through the house wall...

I'm hoping that the systems not too sludged up - after all - it's only been installed for a couple of years......

Agreed with the 'rads out & hosepipe' solution - been there & done that in another property...

Thanks Adrian

[Ed Sirett]

If the system is pumping over then you need to deal with that cause before tackling symptoms.

[Roger Mills]

I don't understand that comment! Pumping over shouldn't cause a change in level - just an unwanted circulation. The only change in level should be due to expansion when the system gets got. The ball valve should be set low enough for the level to rise due to expansion without getting anywhere near the overflow.

[Adrian]

Agreed...

...except this isn't what's happening......

Can't report on the current situation - not been running long enough with the slowed-down pumps - but prior to this - here's what I've observed.

Bale out header tank until it's at the expected level for the float valve (slight pressure on the float causes water to come in via the valve).

Leave system to run. Water level in the header tank gradually rises, until the float valve is completely immersed - then water comes out of the overflow.

Distance between initial level and overflow in the region of 3 to 4 inches..

Have replaced the valve. There is no leak in the float.

The only possibly odd thing I can think of is the thermal store - which is a 200-litre drum situated next to the heatpump. When the system is running, I can hear water running / trickling in the thermal store - According to the absent plumber, the heat store does not have a heat exchanger - so the store contains 'circulating' water.

Not sure if that's a clue or not.....

Any ideas ?

Adrian

[Adrian]

OK - understood.

Pumping over is only an assumption. The water level in the header tank tends to rise, until it overflows. Water in the header tank is warm. Float valve in the header tank has been replaced, and the float does not have a leak...

Any ideas ?

Thanks Adrian

[David Hansen]

On Wed, 05 Mar 2008 20:27:30 +0000 someone who may be Adrian wrote this:-

If water is not coming out of the vent pipe then it is not pumping over.

It sounds like what is happening is that as the system heats up water is expanding up the feed & expansion pipe, which is what it is meant to do. What is the capacity of the F&E tank? What is the total volume of water in the system (include the 200L in the thermal store) and what is the temperature rise? Then use school physics or an on-line tool to see what the expansion of water will be over this temperature rise. Compare this to the volume of the tank.

[Adrian]

Ah well - there's the rub.... The vent pipe and the feed pipe are teed off the same piece of pipe, which appears to connect to the top end of the cylinder coil and to the central heating circuit. Photo at

arrangement - can't quite see what it is achieving...

Thinking back, I seem to recall that the original, oil-fired system used to overflow as well. In the original system there was no independant control of the DHW - the cylinder coil was plumbed off the heating circuit - just like another radiator. When he installed the heatpump, the plumber split the DHW and CH pipework, and put a circulating pump on each. Only one pump is running at a time. They share a common return pipe back to the heatpump. A non-return valve apparently prevents problems (!)

(The F&E tank looks like a standard 'small' one - just enough room for the ball valve... No idea how much water there is in the system, 14 radiators, mix of doubles and singles, plus all the pipework.)

OK. Does the 'expansion' theory explain the observed behaviour, though....?

e.g. System has been running for a while. I go up to loft where tanks are, and bale out the header tank until it's back to the 'correct' level, as indicated by the float. The water in the header tank is warm - say 25 degrees C.

Wouldn't that suggest that any expansion has already taken place ? - and why doesn't the water drop back down to the original level again ?

Thanks Adrian

[David Hansen]

On Thu, 06 Mar 2008 08:38:05 +0000 someone who may be Adrian wrote this:-

Combined feed & expansion and vent pipes are not a good idea. When the work was done it was a good time to provide a separate F&E pipe and convert the combined pipe to just a vent pipe.

You need to draw out how the pipework is arranged if anyone is going to be able to help you further. There are many possible problems.

[Dave Baker]

You need to go and sit by the tank for long enough to see what is happening. If it's pumping over then hot water will be coming out of the vent pipe. If not then it isn't pumping over. Other causes of rises in water level in the CH supply tank are leaks between the CH water and the domestic hot water. Favourite is usually a pinhole in the coil in the immersion tank. This then tries to equalise the water level in the CH supply tank and the adjacent cold water supply tank. Whether you have any/all of these I don't know.

As has been suggested the vent pipe wants to be upstream of the pump where the system pressure is at its lowest rather than downstream where it's at its highest. Sometimes this situation can be created during a system revamp when the new pump is inadvertantly installed the opposite way round to the original one and the vent pipe isn't relocated.

My vent and feed pipes were teed off the same place and I also found my system had been pumping over. Probably had been for years going by the condensation in the loft and the mould on everything stored up there. I just blocked the bugger off with a spare rad valve. The water can expand up the feed pipe just as well I decided if it really wants to. Problem solved anyway.

[Roger Mills]

Looking at your photo (referenced in your other post) you're unlikely to have pumping over with a combined vent and fill pipe - 'cos there's no circulation path. [They should really be separate, down to where they join into the main primary flow pipe].

I'm not sure what the effect of your thermal store is but, ignoring that, if you had a conventional system with a HW cylinder containing an indirect coil, your symptoms would exactly fit what happens when the indirect coil leaks. The domestic hot water surrounding the coil is at a slightly higher pressure than the water in the coil due to the level in the cold water storage tank being higher than that in the F&E tank. If the coil leaks, water flows from the secondary systems into the primary system, raising the level in the F&E tank. I suspect that something like this - if not exactly this - is happening.

[Adrian]

OK - understood....

Hmm - that was one of my thoughts. However..... we replaced the hot water cylinder just after the heatpump was fitted, as the heatpump didn't like the tiny coiled heat-exchanger in the original (cheapo!) cylinder.

So - unless both cylinders had leaks in their coils, it's probably not that.....

I'm still waiting to see if the header tank's filling up at the moment

- to get from 'baled out' to 'overflowing' takes a couple of days ..... so I'll wait to see what's happening this time......

Many thanks for the suggestion Adrian

[Adrian]

Pretty sure that it's not that...... see my other post to Roger ...

From a brief look, it seems that the layout is - pump, hot water tank coil, f&e pipe, return...

I can't see that the water can be finding its way up the vent pipe (which looks pretty redundant at the moment!) - more likely just going back up the feed pipe......

Thanks for the comments Adrian

[Ed Sirett]

A possible cause of this over flowing is that the float level is set too high. The level needs to be just above the OUTLET from the tank. This allows the entire capacity of the F&E header tank for the expansion of the primary water as it heats up. That would be normal operation.