2 cold radiators

Moved in during the summer, now the CH is getting used have found that 2 rads out of 10 are cold.

System - is fully pumped single pipe (3/4" or 22mm) in older part of house, or twin pipe 15mm in new part. All rads have TRVs apart from the towel rails. Combi boiler ca 8 years old therefore no header tank. Some previous twit has fully opened all the lockshield valves so it will need rebalancing once I can get two of the rads working again.

Problem - Two rads, both on the older single-pipe part of the system are cold. I have tried turning all other lockshield valves off to direct maximum hot water just at these 2 cold rads but that doesn't help either.

Cold rad #1 just gets luke-warm along the top, rest is cold. It has the LSV on the flow, and Danfoss TRV on the return. The flow pipe gets hot up to the LSV. The return with TRV on it is cold. The Danfoss TRV is the type with the grey plastic screw collar, it's bi-directional, and the arrows indicate that it has been correctly set to be on the return side. I've had the head off, the pin moves in and out ok. I've turned the LSV fully in and out several times, so without taking it off, I assume that is unrestricted.

Cold rad #2 just gets very luke-warm along the top, rest is cold. It has an older Danfoss TRV on the flow, and the LSV on the return. The flow pipe is cool up to the TRV. The return with LSV on it is cold, but the short return between the LSV and the main single-pipe return is hot presumably because it's being heated from the main return. The Danfoss TRV is the type with the black plastic push-on collar, secured by an Allen Key, it's directional, it's been put on the correct way (flow). I've had the head off, the pin moves in and out ok. Again I've turned the LSV fully in and out several times so without taking it off, I assume that is unrestricted.

I have bled all rads including these two - no air appeared, only water. No tell-tale blackish water, only pure and clear.

Questions:
1. Could it be that the pins in the TRVs move freely but the valve inside is stuck? If so can that valve be freed without a drain-down, or does it mean new TRVs?
2. Is it more likely to be sludge? If so I'll take the rads off in the Spring. Any suggestions for a desludge product that works, and the best way to use it?
3. Or is it more likely to be an airlock?

Thanks!

With the one pipe system, the pump pushes water around a loop of pipe. Because of this, the pressure differential across a radiator inlet and outlet is very small as they are both connected to the same pipe receiving the same pressure from the pump, so flow through them is slow, mostly by the gravity principle of hot water rises, cooler water falls. All radiator valves on a one pipe system should be fully opened to assist this process. Because the radiators sit atop the pipe, and the flow and return are both into it having the valves open makes no difference to the pressure at other radiators on the loop.

First of all, is the one pipe "loop" of pipework hot along its entire length, or is it cold at one end? If it's hot the pipework is OK.

If the pipe isn't getting hot, is there a manual by-pass valve somewhere that has been opened too much allowing the water to short circuit the radiators? or perhaps there is a balancing valve somewhere on the one pipe circuit that has been closed down too much?

If the pipe is getting hot, then the problem is that the water is there, but not circulating through the radiators. Try taking one off and flusing it through to see if it's full of sludge. The sludge is heavy, so it sits at the bottom.

Are the TRV's fitted suitable for a one pipe system? There are special valves for one pipe / gravity systems that have greater flow than the standard valves to allow for the lack of pressure differential. Because of this lack of pressure there may not be enough to operate normal TRV's. (some valves, Terrier I think, stiplulate that they are for two pipe systems only)

Having said all of that, maybe the real answer is to convert the one pipe system to two pipe and then balance it up properly.

Thanks stem, that is really helpful. One-pipe is new to me, so I'm learning - good explanation!

Pipe warmth:
- the main one pipe 'loop' around that part of the house is hot all along its length;
- on one of the cold rads the flow 'spur' is hot and the return 'spur' is cold;
- on the other cold rad the flow 'spur' (according to the Danfoss TRV orientation) is cold and the return 'spur' is hot (but I will investigate further because I can't see all the pipework there, the flow and return might be the other way round, in which case it is possible that the Danfoss TRV on that one may have been fitted against the flow).

By-pass/balancing valves on the one-pipe system:
None visible.

I hadn't realised that some TRVs are unsuitable for single-pipe. Will investigate further. On the rads on the single pipe part there are 2 Honeywell TRVs which work ok; 1 Oventrop TRV which works but that rad never gets really hot; and 3 Danfoss TRVs all different models (and 2 of those are on the cold rads).

It does sound like it might be sludge, so in the Spring I'll take the 2 cold rads off, flush them anyway, check or replace valves etc.

The other thought I had is that the 2 pipe part of the system is taking too much flow. because it always heats that up very quickly. Both the cold rads are towards the end of the single pipe loop and the flow will be cooler at that end, so the gravity effect is presumably going to struggle more (although it does seem to be hot on the main pipe loop there). There is a service or isolation valve (22mm compression, quarter-turn to open/close, using screwdriver in slot) on the flow to the 2 pipe part of the system near the boiler. If I close that a bit then presumably that would throttle the flow to the newer 2-pipe and improve the flow to the older 1-pipe part? Do you think that the single-pipe flow would increase if I did that, and if so can service/isolation valves be used like that (they seem to leak as soon as they are used!), if not are there valves which are made for the job? I can't do it yet because a CH manifold valve (flow) is leaking slightly on the boiler (Worcester 35Cdi) and I don't want to stress the system until it's replaced.

At the moment I'm keeping the pressure at 1 bar and it seems to stay at that for weeks before needing to top-up, even with the leak. When the manifold valve has been fixed, do you know if these pressurised systems work better with a higher pressure than 1 bar, or does it make no real difference?

But having had your comments - thanks again - I think I now have a logical plan in mind to get to the bottom of this once the weather gets warmer, and if all else fails then it may be that I'll need to add a second 22mm loop and convert to 2-pipe but that would be a major project.

daverk said:

The other thought I had is that the 2 pipe part of the system is taking too much flow. because it always heats that up very quickly.

Click to expand...

Two pipe systems will heat up more quickly because the full pressure of the pump is directed at the radiator inlet & is pumping the water actually through the radiator. On the single pipe it just flows across the bottom of the pipe leading to the inlet, leaving nature to do the rest, slowly.

daverk said:

Both the cold rads are towards the end of the single pipe loop and the flow will be cooler at that end,.

Click to expand...

One pipe systems do get cooler towards the end of the run as the previous radiators deposit their cooler water back into it cooling it down before it reaches the next one. A properly designed one pipe system can have larger radiators or more of them at the end to take this effect into account. If the loop is hot all the way around, I don't think flow is your problem. The flow in a one pipe is the same all the way around, the same quantity that goes in one end comes out of the other, it's just a bit cooler. It only cools slightly from one radiator to the next, and as the TRV's on the earlier ones shut down more hot water is directed to those further down stream. If one radiator gets hot and the next one is cold despite the "one pipe" it's sitting on below being hot, then the water is reaching the radiator but not flowing through it for some reason.

daverk said:

By-pass/balancing valves on the one-pipe system:
None visible.

Click to expand...

Doesn't have to be on the one pipe part of the system, it could be anywhere that water is flowing around the system. Having said that as a "one pipe" loop is a by pass in itself in that it that water is continually circulation around it.

daverk said:

so the gravity effect is presumably going to struggle more.

Click to expand...

The temperature difference across flow and return doesn't change that much from radiator to radiator (assuming they are similar sizes) as the inlet temperature is lower than the previous radiators inlet temperature, the outlet is also correspondingly lower. Gravity circulation depends on the difference between flow & return.

If the entire loop is 22mm there shouldn't be too much resistance there and so I imagine you would get a fair flow though it.

Higher pressure of the water in system overall won't make any difference to the flow rate, as the higher pressure will be on both the flow and return parts of the circuit, and effectively cancel each other out.

As an after thought. One pipe systems work best when the radiators sit directly above the circulation pipe, if the circulation pipe is to the side of the radiator with an interconnecting branch on a not very steep angle, it reduces the effectiveness of the circulation around the radiator.

stem said:

As an after thought. One pipe systems work best when the radiators sit directly above the circulation pipe, if the circulation pipe is to the side of the radiator with an interconnecting branch on a not very steep angle, it reduces the effectiveness of the circulation around the radiator.

Click to expand...

You may well have hit on a contributing cause of the problem. Both the cold rads are off to the side of the main 22mm single pipe (at places where it is vertical), with nearly horizontal 15mm flow and return 'spurs' to the rads. ( All the rads which work sit above the single pipe). If there's no sludge, and the TRVs prove suitable for single pipe, another option might be to try altering the angle of the flow/returns by moving the rads a little (I assume that these rads did actually work once - perhaps the brackets have slipped a bit). In fact, I think I'll try that first. If that doesn't work, perhaps replacing the 15mm spurs with 22mm might help. But that's for the Spring.

I hadn't realised before your clear explanations that pumped single pipe systems worked on the gravity principle once near the radiator, although it's logical when you think about it!

Maybe they've never worked. I'm sure there is a recomended minimum gradient, but the steeper the better, but one thing's certain, horizontal runs won't work. Sounds like you've found the problem.

BTW I've got a 1960's single pipe feeding 7 radiators with a two pipe 2006 extension feeding 3 radiators. It works perfectly, however the hall radiator pipes originally went through the wall to connect to the one pipe in the lounge, and whilst it got warm at the top it never did at the bottom. I lifted the radiator and increased the angle. It's hot all over now.

Thanks stem. We've certainly narrowed it down a bit. I'll post again when I get them working again, but that may be a while.

Just in case this might be helpful to someone else with a single pipe system (or part single pipe) here's the outcome to the problem I first posted here last year:

- there was no sludge in the rads
- they all had standard Thermostatic Radiator Valves, which are designed for 2-pipe systems, so gravity flow through the rads was probably restricted a bit
- on the rads which were cold, the TRVs had been fitted in each case on the connections furthest from the boiler (i.e. the return ends), and the flow was going, or trying to go, 'backwards' through them!
- one radiator was plumbed so in was trying to discharge against the main single pipe flow, so it couldn't work at all

So what I did was, on the single-pipe system:

- took the rads off and flushed them through anyway
- reconfigured the pipework so that where possible the flow connection was at the top of each radiator (supposed to improve gravity flow through each radiator) rather than at the bottom , and ensured that all returns could rejoin the single pipe easily and not fight against its flow
- fitted large-bore Danfoss RAG-15 TRV bodies with RA2910 sensor heads (unused ones off Ebay) to all the radiator inlets, with Danfoss RLV large bore lockshield valves on the returns (Danfoss tech support is very helpful)

With that, I got heat flowing to all radiators again, but some were a bit luke warm.

Finally, I turned up the boiler temperature a bit and that got very hot water flow to all the radiators, and they have all been working well over the winter. (In fact, the higher boiler temperature increases gas consumption by only 3%).

I think that once upon a time the system had been single pipe with standard non-thermostatic valves, which was probably ok, then someone fitted TRVs which were off-the-shelf 2 pipe ones. And there was an additional complication because I think that a new boiler had been fitted at some stage, and the original single pipe flow direction had probably been reversed, hence the reversal of flow through some rads where TRVs had already been fitted (although I think at least one had actually been fitted backwards - there was a £500 bill I found from a plumber who had fitted one rad with valves, backwards, for a previous occupant!)

Hope that helps someone. The cheapest solution would probably have been to switch to 2-pipe, but that would have been more disruptive.

Thanks for posting the update. Good to hear how things turned out, and that everything is working OK.