Back feed on central heating

[Chris Williams]

About 2 years ago I had my boiler moved and an additional radiator installed, since then the radiators have been getting warm through the returns when only the hot water is on. The only changes to the original pipework are as follows: The return for the new radiator tees off past where the central heating return joins the hot water return, yet this radiator does not get as hot as the rest of the radiators. This one uses microbore as opposed to the 15mm of the rest of the radiators. The flow and return are now longer but the original pipe work has not been altered before these extensions,other than above. One other point is there is a pipe teeing off from the flow between the pump and the three way motorised valve that goes through some sort of calibrated valve and then is connected back into the return, now this pipe gets as hot as the flow immediately the hot water comes on, is this right? I have British Gas cover but I am afraid the plumber might have inadvertently done something wrong and they might not cover this! Help!

[Andy Hall]

This one is a classic. The usual reason is that the places where the flows split and returns rejoin are not at equivalent points and a flow is induced where it shouldn't be.

If you map out the pipework on paper and put in the places and lengths involved, it will become obvious.

This could be partly or wholly responsible for the phantom flow effect as well.

You don't say what size or capacity of radiator it is or the pipe lengths and sizes. Microbore comes in 6mm, 8mm and 10mm (occasionally 12mm) and is fine as long as adequately sized for the radiator over the distance involved. 6mm will deliver enough water to support about 1500W over a few metres, 8mm 2500W and 10mm about

3500W. If you could measure the radiator height and width in cm and also how many panels and sets of fins it has, it should be possible to estimate its capacity and whether the pipe is adequate, given teh lengths of the microbore as well.. It is also possible that the system hasn't been balanced - some plumbers are lazy. Before tweaking anything we should do a sanity check on the pipes because if they are grossly undersized, no amount of adjusting will fix the problem.

This one is an automatic bypass valve and yes it should behave like this. It is there to dump heat from the boiler when the valves close and it has been firing - it prevents it from tending to boil.

British Gas cover isn't worth the paper it is written on, let alone the money. I doubt whether they would fix a problem as a result of a badly done modification for free If there is something wrong then the plumber should come back and fix it free of charge,.

.andy

To email, substitute .nospam with .gl

[Chris Williams]

Chris Williams replied:

Thanks for your help Andy as you might guess I am no heating expert! When you say that where the flow splits and the returns join are not in equivalent places do you mean that they should be the same distance from the boiler? This is the only distance that has changed as where these are now in relation to the system before the alterations they have not moved. I am also guessing that the flow splits at the 3 way valve. I think I might have misled you regarding the additional radiator when I said it did not get as hot as the others I meant with this back feed problem. I have the figures anyway just in case:

10 mm Microbore; Flow is 315 cm approx. and the return is 120cm approx. The radiator is a single panel with no fins and measures 40 cm x 60 cm (w x h). Could I put a non-return valve on the central heating return and move the return for the new radiator to the other side of this valve? Thanks again, Chris

[Andy Hall]

OK, Chris. What should happen (typically) is the following. It's not the *only* solution, but this is the text book one:

- Flow from boiler has feed and vent pipes (assuming you have an open vented system) connected to it and then goes to the input side of the pump. _ output side of pump goes to 3way valve or zone valves. One branch goes to cylinder, other to all radiators. Obviously the branches off to the radiators all leave after the 3way valve. The critical part is on the return side. The returns for the radiators must all join the return pipework *before* the return from the cylinder joins it. If any radiator returns join the return path

*after* the cylinder return then a reverse flow can be created through them when only the cylinder path should be active.. This can be diagnosed by feeling what should be the return pipes of the affected radiators. They will be warmer than the normal flow pipes.. the fix is to make sure that all of the radiator return pipes join a heating return before the return from the cylinder then joins. It's a bit hard to figure out from your description as to whether this is what you have, but it sounds like it. The new one is probably not getting as hot because the flow through it is smaller anyway. The new radiator itself is a fairly low power one and I am sure that 10mm is fine You shouldn't need to use non return valves. Fixing the order of the pipework should do it.

.andy

To email, substitute .nospam with .gl

[Chris Williams]

Andy replied; OK, Chris. What should happen (typically) is the following. It's not the *only* solution, but this is the text book one:

- Flow from boiler has feed and vent pipes (assuming you have an open vented system) connected to it and then goes to the input side of the pump. _ output side of pump goes to 3way valve or zone valves. One branch goes to cylinder, other to all radiators. Obviously the branches off to the radiators all leave after the 3way valve. The critical part is on the return side. The returns for the radiators must all join the return pipework *before* the return from the cylinder joins it. If any radiator returns join the return path

*after* the cylinder return then a reverse flow can be created through them when only the cylinder path should be active.. This can be diagnosed by feeling what should be the return pipes of the affected radiators. They will be warmer than the normal flow pipes.. the fix is to make sure that all of the radiator return pipes join a heating return before the return from the cylinder then joins. It's a bit hard to figure out from your description as to whether this is what you have, but it sounds like it. The new one is probably not getting as hot because the flow through it is smaller anyway. The new radiator itself is a fairly low power one and I am sure that 10mm is fine You shouldn't need to use non return valves. Fixing the order of the pipework should do it. /andy Chris Williams replied; Thanks again, Andy, I will try it and let you know, might not be for a while though as I intend changing bathroom and re-siting the radiator in the Autumn so might leave it until then!

Chris

[Ed Sirett]

Taking account of the locations, (in)accessibility of pipework, likely damage to decor, etc.... A non-return valve might be the simplest course of action. If so it will be 99.9% certain that the radiators will need to be balanced after installation of the valve.

[Andy Hall]

Although I suppose that one would need to be very careful about location of the NR valve to make sure that it is not in the path from the boiler to vent facing the wrong way to allow steam to escape in the event of a fault??

.andy

To email, substitute .nospam with .gl

[Ed Sirett]

Well, it can go in the return where there should be no issue, your point is neverthe less quite valid.