That time of year...
Gas fired CH. Pumped system. Three-way valve for hw/ch/hw&ch. New pump installed last year.
Problem is that CH isn't circulating. It starts up okay, and the radiators get hot. But then circulation stops, even though the pump is still running. So the boiler short cycles, shutting down with the internal stat. Then, randomly, the circulation will suddenly start up again, with noises of bubbles, so I'm guessing there must be air in the system.
I can't see an automatic vent in the loft, where the pump is.
When I replaced the seized pump last year, I took it apart and found black haematite, so there may be sludge in the system. The TRVs I've looked at seem stuck open...
So I'm guessing airlock or sludge somewhere. I've bled the radiators, but they're all clear, so any airlock must be higher in the system.
So, what should I be looking for in the loft as a bleed mechanism? And are there any other suggestions for possible causes?
Fernox F3 flushed purchased and will be left in for a week, then I'll drain the system, replace all the TRVs, and refill with F1.
All advice welcome, thanks.
Take all radiators off the wall, take them into the garden and flush them through with a hose.
That would sort sludge. Do you think the intermittent behaviour is caused by sludge?
Try running the system and then turning all the rads off bar one. Then crank the thermostat up. That tends to push all the air into that one radiator. It worked for us when we had air in our system.
I had similar this time last year....It was the “dead spot” or neutral pressure point in the pipe work that was blocked. You should have an “H” pipe arrangement just prior to the pump.
in mine the cross bar of the H piece was badly blocked. Took a hammer and long screw driver to break up the blockage ( I too had previously tried fernox and flushing)
> Take all radiators off the wall, take them into the garden and flush them through with a hose.
It would be a lot easier to get the system flushed with the rads in situ. And not ruin your carpets!
If it is a blockage, I understand that they can occur at the junction where the feed from the F&E cistern joins the system, just before the pump.
Forgot to mention that the f&e tank overflow is dripping, too, so that might point to some sort of blockage you describe.
Possibly, although that sounds to me like the float valve not shutting off properly.
Discovered the actual cause of that drip; hot water tank header tank. Not the heating cistern.
Mashed the float valve a few times, and bent the arm down, and it seems to have stopped dripping. Will get a replacement valve and float, as its ancient... At least the stop valve works...
Found a couple of air bleed valves; one on the hw cylinder, and one on the pipework near the pump. Neither vented very much at all.
Immersion heater isn't working, so chased the electrics as far as the switch box, which is dead. Not sure why, as that goes straight to a dedicated isolator on the CU, which isn't tripping.
Not a very successful day...
If the heating is going off and then coming back on it sounds like a faulty room stat .
> faulty room stat
Heating doesn't go off; pump still runs, boiler still fires. Just the water doesn't circulate. Room stat is fine.
If the room stat is faulty or is satisfied, up to temperature.it will close the heating port on the 3 way valve. If you have a low water content boiler the boiler controls the pump, keeps it running until it's cool enough to shut down. It's called pump over run
The three way valve doesn't change position. If I fiddle with the stat, I can hear it trigger (click; it's old skool bimetallic), and that turns the boiler on and off immediately.
If the stat wasnt working, the boiler wouldn't keep short cycling; if the stat was open, when it should be closed, it would keep the boiler turned off.
ps. my system isn't smart enough to do pump overrun; it's an old boiler, so it can cope with the thermal overrun, with a tiny bit of kettling, and then thermosiphon.
Does the valve change position when you switch between hot water and central heating?
You should be able to take the actuator (head) off and see if the valve operates using a spanner. You’ll then also see if the actuator actuates. Maybe you’re getting warm radiators due to conduction, or the valve is only opening slightly.
> Does the valve change position when you switch between hot water and central heating?
Yes, it seems to work normally. At least, the indicator moves position. I'll have a look at the actual valve, rather than just the actuator, though. But the fact that the system circulates to start with suggests the valve is working okay.
> Maybe you’re getting warm radiators due to conduction, or the valve is only opening slightly.
No, when the system starts up, the radiators get properly hot, and both in and out pipework gets hot, and I can hear water circulating.
When the circulation stops, the inlet pipe is hot, but outlet is cold. I'm a bit confused by this behaviour, because it's not thermosiphoning down... And the inlet pipe can suddenly get cold.
Maybe run the pump on lowest setting. Maybe you’re somehow dragging cold water into the system from the header tank. The bubbling might be hot water expanding into the header/expansion tank. I don’t see how the outlets can be ‘cold’ with the inlets hot. Unless maybe you have TRVs on every single radiator. One radiator should be always open and not have a TRV on it. Usually a bathroom or hallway one.
What make/model of boiler do you have? Is the divertor valve an old H/well ? If over the years the system has had a mixture of different chemical treatments I've known this to cause the rubber ball into valve to swell up when it gets hot. I've only seen this twice, I'm a retired heating engineer. The rubber ball will be soft and sticky not hard.
P.S. Thermo siphoning is usually referred to as gravity circulation when talking about domestic heating systems.
> Maybe you’re somehow dragging cold water into the system from the header tank. The bubbling might be hot water expanding into the header/expansion tank
The tank isn't hot, and there's no overflow recirculation back into the tank, so it can't be pulling water into the system. It lasts for a long time, too, and I can hear that there's no circulation.
> I don’t see how the outlets can be ‘cold’ with the inlets hot. Unless maybe you have TRVs on every single radiator.
I don't understand that either... I've taken the TRV control heads off. The valves do seem stuck, and I plan to replace them all. Bathroom doesn't have a TRV.
Is it a closed system with a pressure gauge on the boiler then? Maybe there’s not enough pressure in the system?
There must either be a feed and expansion tank or a filling point.
> Is the divertor valve an old H/well ?
It is an old Honeywell, yes... It's been replaced before: the old one is still up there from when I moved in... Does the ball swell so much it stops circulation in the position it's in? Rather than not moving into position? It's moving into position correctly, unless the actuator can change, but the valve doesn't.
Boiler is an old Netaheat(?). Still seems to be working okay.
> Thermo siphoning is usually referred to as gravity circulation
I know, but that's usually used when you have gravity circulation as the system design (e.g. pumped CH and gravity circulation HW) mine doesn't: both are pumped with a 3-way valve.
It's not a closed, pressurised system.
> There must either be a feed and expansion tank or a filling point.
There is, but, in order for it to pull cold water into the system, the hot water already in the system would have to go somewhere. I've seen some diagrams that show an expansion pipe leading back up to the feed/expansion tank. But mine doesn't have that: only two pipes; cold feed at the top, and output to the system at the bottom. Plus overflow out.
As I said, I can hear that circulation has stopped; it's not a matter of circulating cold water, from wherever. It's just not circulating...
If I remember correctly the Potterton Netaheat has a pump overrun. It has a cast iron low content heat exchanger. Take a look at the old valve there is not much room inside, the ball can swell and become loose. Sounds like you should get someone in to look at it. Yours is not a straight forward problem.
I have to figure out the heating electrics; it's a shambles of wires and individual choc block connections. I've looked at it for years, and it offends me, but I've always taken the approach of "it ain't broke, so don't fix it", and left it be. But, as an electronic engineer, a bit of simple electrics shouldn't faze me. But the installation manual for the programmer is gibberish. I'm getting there, though...
The point being that figuring out the electrics would help understand exactly what controls what; I'm not sure if the boiler is involved with the pump control; I think that's just programmer and stats.
I may have to resort to paying someone...
Just count the cables coming from it.
Should have power, stat, 3port valve etc. You don’t have to identify which ones are which, just how many there should be.
> Just count the cables coming from it.
Done that. Which is why I suspect the boiler isn't controlling the pump.
> Should have power, stat, 3port valve etc. You don’t have to identify which ones are which, just how many there should be.
If I'm going to sort out the wiring shambles, I need to know what is what, so I can connect it properly. And so I know what is controlling what.
Does the system drain and fill up normally via the header tank?
Sounds like you either have an airlock or a physical blockage of the system.
When did you last run the central heating ?
Do you use the boiler to heat a hot water cylinder and is that running ok?
You should be able to download a Honeywell Y plan diagram and a boiler wiring diagram.
If all has been wired correctly at the boiler you should have P. Live in. S/l in from Orange at the valve to fire up. S/L out to the pump. Cyl and room stat are usually connected to a wiring box near the valve. But some times everything is wired back to the programmer, that can be a headache.
It might be easier to disconnect all the wiring and reconnect it in order rather than try to figure out what's wrong. Good luck with that.
> Does the system drain and fill up normally via the header tank?
I haven't drained it yet.
> Sounds like you either have an airlock or a physical blockage of the system.
Any easy means of determining which?
> When did you last run the central heating ?
Earlier this year. Been running it since early October, but I think the problem was present to a lesser extent previously. I replaced the pump last October, as the last one seized. Having cleaned out the black haematite, it rotates freely again. So there may well be a haematite blockage somewhere in the system.
Is it possible for the pump replacement to have introduced an airlock? It has isolating valves either side, so didn't need the system draining, so it would only have introduced air to the volume of the pump, which isn't much (mostly in inlet and outlet ports).
> Do you use the boiler to heat a hot water cylinder and is that running ok?
I do. Whilst I do get hot water, it is sometimes not as hot as normal, so I think the hot water circuit isnt working perfectly, either.
> You should be able to download a Honeywell Y plan diagram and a boiler wiring diagram.
Yes, I've found many of those. I'm interpreting them with an electronic engineer's eye. I need to understand the function of each component for the schematics to make sense. Getting there, slowly...
> If all has been wired correctly at the boiler you should have P. Live in. S/l in from Orange at the valve to fire up. S/L out to the pump. Cyl and room stat are usually connected to a wiring box near the valve.
I'll have to figure out your terminology...
P.live = permanent live?
S.live = switched live?
> But some times everything is wired back to the programmer, that can be a headache.
All the connections are in a rat's nest of wires in the loft. The programmer and boiler are downstairs. Makes for tricky diagnostics...
> It might be easier to disconnect all the wiring and reconnect it in order rather than try to figure out what's wrong.
That's what I'm planning to do, at some point. But I'd like to make sure I understand how it's currently connected.
Not helped by the installation using bog-standard three core flex for connections between programmer and loft, so the wiring looks mad: blue to brown or Y/G, etc. Plus the 3-way pump wiring colours.
I think I've assembled all the documentation, so it just needs me to sit down and think it through carefully. Blindly following a diagram I've found on the internet isn't my sort of approach; I need to understand what's going on first. As it's all mains voltages, mistakes could be unpleasant...
> Good luck with that.
BTW, whilst my replies might seem rather 'matter of fact'; I'm grateful to all respondents for taking the time to reply with advice; thanks everyone.
One thing you could try is to simply shut off all the rads bar just the one...see if that somehow keeps the water moving through the system enough to keep the boiler running...
Taking the actuator body off the Y valve and manually opening it worth doing of course.
But it does sound like a blockage and my money would be just prior to the pump, around the area that the F&E pipes join the pipe work.
> But it does sound like a blockage and my money would be just prior to the pump
A blockage would make sense. Apart from the fact that it all seems to work for about half an hour, and then stop circulating. I'm struggling to think of a mechanism for that.
Unless the blockage causes an airlock...
The pump, being modern (Grundfos UPS3), and electronic control, knows when it's blocked. I don't think I've seen the red blockage light come on. I need to monitor that better, though.
It’s all very frustrating...I was in same boat same time last year and kept putting off the inevitable (it was reasonably warm at the time) eventually bit the bullet and drained the system and got the pipe cutters out.
It is frustrating and a bit disheartening; I think I should be able to sort these problems...
I spotted a nice tip on the hot water thread for detecting blockages; since it's black magnetite (not haematite as I said earlier), it might be possible to find blocks using a magnet. Any magnetite blockage will attract the magnet. Will give that a go...
Very rare ( only 2 in my career) and very random in its effects Is the ball inside the 3way shears-off on it’s brass arm And then randomly moves around the inside of the valve body sitting in different positions regardless of the correct actuator function.
Easy to prove once valve removed From the pipe work as you can shake It and see/ hear the ball fall about. If your Honeywell actuator has a round minstrel sizes bobble in the metal top of the actuator it means you can remove the actuator from the body without draining the system. Beware older ones without this do not come off without draining/dismantling.
loft sounds like a poor position for a pump assuming standard house layout as it will have little head of pressure from the F and E in the same loft?
if the system worked historically and nobody has messed with the wiring it is unlikely to be the issue.
very badly sludged systems can do various strange things with circulation. Sounds like you have a combined feed and vent which I think the old Pottertons allowed to make for easier pipe runs but it is definetly a poorer system than a close coupled separate feed and vent pipe arrangement and far more likely to sludge up.
I think the issue is mechanical rather than electrical.
They tend to be wired by plumbers so usually a mess. The problem is if you disconnect it all and the reconnect it, and it doesn’t work, you now have two problems.
Really sounds like you’re whole system is sludged up. Bits of sludge are moving and you get a rush of water when they move and then it all stops again. Find the drain point and drain and get a flush done. I’d pay a plumber who has all the flushing equipment. Rather than mess around tying to get all the bits and pieces, cause a leak and ruin a carpet. 😂
Your system could very little water in it .
Please try to drain it and see if the header tank tries to fill it up .
Could be a blocked feed pipe , and your heating system is drying out even though the header tank is full .
please don't delay as this can do a lot of damage
> if the system worked historically and nobody has messed with the wiring it is unlikely to be the issue
Oh, I know; the wiring is a side issue. It just offends me. And it's a displacement activity, probably...
I'm assuming it's mechanical.
The radiators are all full and they heat up.
> Your system could very little water in it .
The vent ports at the top of the system very quickly produce water when opened, so I don't think the system is running dry. Thanks for the warning, though. I need to drain the system a bit to stick in the flusher, so that will tell me if the infeed is blocked.
> cause a leak and ruin a carpet
Carpets are shit, anyway...
Glad your system has water in it .
Either an airlock or blockage to look for now
Sounds like what we had !! we finally just got the plumber to get rid of all the sludge in our system (haven't had a the bill yet, but I would expect quite a lot! Two of them from 8.30 -5.30 running up and down totally emptying and refilling the system, you should have seen the sludge ...disgusting! The result is heavenly, quiet, no noise and good heat from all the radiators! Really happy with it (and that is after I tried all the radiator bleeding, etc my self... which was not not really working)
Mountaineering Scotland is urging hill goers to prepare for the clocks going back this weekend. They are asking every person in the hills to pack a headtorch - and a spare - as British summertime comes to an end.