Ineffective central heating pipe design??

[Alister21]

I have a recurring problem with the radiators on the upper floor of my house. From experience I have de-sludged and inhibited them in late autumn but still the upper system chokes around this time of year and the radiators go cold. On the ground floor all pipework is 22 mm and 15 mm copper (no problems there). However, on the 1st floor it is 22 mm copper to and from a manifold and then 10 mm microbore. My problem is this. Am I right in thinking that the flow and return lines should serve two seperate manifolds? My system has them both flowing through one manifold with the 10 mm microbore flowing and returning from it. Is this right? I would have thought that the water would follow the easiest route i.e. by-passing the radiators and flowing straight back to the boiler. I think that as soon as I get any sort of sludge in the microbore then that is exactly what is happening. Every radiator in the house has a thermostatic valve, not a lockshield in sight. No sign of a PRV anywhere else in the system either. I take it from that, that the single manifold in question is not some sort of internally divided job and it is 'straight through'. Coming to the point now....Should I split the line here and install two seperate manifolds and some sort of pressure relief between flow and return?

[Andy Hall]

I suspect that you will find that this is actually a manifold with two separate flow and return sections.

Try measuring or feeling the temperature of the pipes feeding it on the boiler side and those going to the radiators. If it really were going in and out as you say, there would be little or no flow through the radiators, even when they are clean.

Does everything work OK immediately after cleaning? If not, then there is probably a balancing issue and the radiators on the 15mm sections need to have their flow reduced.

Certainly there should be lockshields on all radiators even if TRVs are used, or the radiators on 15mm connections will probably receive most of the flow until those rooms are warm.

If it deteriorates after a while and that is sludge, then there is something else wrong. If you are using inhibitor, then it should last for at least a year assuming you are using enough. There should be little or no sludging. Even 8mm microbore works perfectly well over long periods of time.

Suspects for early failure of the inhibitor and sludging would be pumping over or sucking down of air in the feed/expansion tank. Have you checked this carefully? Pumping over is obvious. Sucking down will usually cause a sound of air bubbles being pumped around and it will collect in the radiatrors.

[Guy King]

The message from "Alister21" contains these words:

You sure it isn't one block with two passages, one flow, one return?

[Alister21]

Thanks Andy. One of your points has me thinking. This 'sucking down' thing. I once cracked the upstairs bleed valves with the pump running (bad practice I know). The needle valves sucked in air rather than released water. Is this syptomatic of the problem you quoted or would you expect that to happen every time? And if it is a sucking down problem how can it be remedied / prevented?

Al.

[Fred]

Generally there can be 3 problems

1 Pump-over during normal running - disaster!

2 Pump-over when pump starts.

3 Pump-over when the pump stops.

I'd have said "suck down" is less likely because of the head of water which has to be overcome for it to happen.

Any of the 3 will aerate the water with obvious consequences.

[Alister21]

Guy, I have never broke the line there as I did not want to disturb all those 10mm joints. If memory serves (two winters since last I looked) it is a cast unit and may well be internally divided. Andy suggests that I feel both sides for a temperature drop and I will probably do that next weekend. Has anyone ever came across a manifold like this? Al

[Andy Hall]

That is concerning because the system should always be under positive pressure, even with the pump running. You could have air being sucked in through a radiator vent somewhere, for example.

Is the Feed/expansion tank in the loft, one floor up from the highest radiator?

Can you do a description of the relative positions of the boiler, pump, motorised valve(s), the feed/expansion pipe, the vent pipe, where the returns from the CH and cylinder coil connections join etc.?

It's not uncommon for items to be placed in the wrong order. A classic is for the FE pipe and the vent pipe to be on opposite sides of the pump or separated by more than 150mm or so of pipe - e.g. opposite sides of the boiler. That's the kind of thing that can cause the pump over/suck-down issues, although sometimes they are subtle.

If you have a way to do a sketch and post on a web site somewhere it would be ideal. The more information the better.

[Alister21]

Excuse my ignorance lads but please explain 'pump over' / its causes / its cure.

Thanks for all replies to date. Al.

[Andy Hall]

Yes.

I've seen them with a 22mm compression fitting at each end and a different type with both 22mm fittings at one end with branches on each side.

They don't seem to be so easily available any longer in these designs, possibly because of the cost. More typical are brass blocks with 3 or 4 holes for 10mm or 8mm tubes. These are inserted into a 22mm fitting and the lot soldered.

e.g.

[Alister21]

Andy I will need to do some scrambling about but basically the boiler is on ground floor (kitchen). The pump is under the ground floor - I think there is a motorised valve down there as well. The hot water tank is on the first floor. The expansion tank is in the loft one floor above the highest radiator.

Sorry it is very general but as I say the rest I will need to refresh my memory with regard to the rest.

Al.

[Andy Hall]

See other post.

Basically it normally results from the pressure head between the FE and vent pipes being too high because they are wrongly positioned. However, there are scenarios where other places can be under negative pressure as you have seen.

[Andy Hall]

OK.

I suspect that the devil might be in the detail here, but from what you've said, I think that we are looking at air getting into the system if it's OK immediately after cleaning and then deteriorates in a few months.