Googling this issue I'm finding lots of conflicting answers and many horror stories about significant parts of the system needing replacement, so I'm trying here for, hopefully, a more considered response.

At the weekend I replaced a couple of sticky TRVs. To do this I drained the central heating system the same way I have done a dozen times in the past (to replace rads, replace TRVs, replace valves, fit filters etc).

I refilled the system as normal and bled the rads a couple of times starting with the lowest rads and working my way upstairs to the highest rad (a heated towel rail).

All seemed normal except that there seemed to be more air in the system than usual.

Over the last two nights I have bled the radiators again and found a lot of air in a couple of them again. Other than that the heating seems to be working as before.

However, when the heating switches off at the end of its time cycle (and possibly when the room stat kicks in - though I haven't confirmed this) water pours out of the header tank overflow pipe for a few minutes then stops.

This seems to happen every time the heating switches off.

System details:

  • Worcester Bosch 24 CBi (I think that's the correct model) boiler
  • Osso unvented water tank
  • Grundfos pump
  • TF1 magnetic particle filter
  • Radiator circuit is fed from one small tank in the loft
  • Cold water into hot water tank is fed from mains feed (via PRV) - not a big tank in the loft space.

The boiler is over ten years old BUT had the burner, heat exchanger, fan etc (everything except the case and control board, basically) replaced two year sago.

The Grundfos was replaced at the beginning of this year. (It was replaced needlessly by incompetent heating engineers who claimed the previous pump was not powerful enough as they had added an extra rad to the system. They were ignorant of the fact that two larger rads had been removed from the happily working system AND the fact that the 'fault' was that rads were not heating up because they had left the lockshields closed!!!)

Other than the pump change and new rad at the start of the year, nothing has changed.

The system has worked perfectly since then (after I discovered the closed lockshields and the two cross threaded TRVs left by the supposedly certified heating engineers).

When I changed the TRVs at the weekend, the system drained down as normal and re-filled as normal.

This overflowing from the header whenever the heating switches off is the only problem. Everything else is working fine.

What I have read and considered:

  • Boiler over-heating. Seems unlikely as the water temp at the taps is not abnormally hot - it does not seem hotter than normal.
  • Ball valve in header tank stuck. Seems unlikely as the overflow stops itself after a few minutes - and there is no sound of water pumping into the header tank when the overflow happens (I need to confirm this again, but when the overflow first happened I was in the room where the header tank is above and usually hear that filling).
  • Tank stat faulty. I know a tank stat is supposed to be fitted, but I don't think one is. Heating engineers have asked where the tank stat is before and not found one. Again this is something I need to check, but is there any way to further diagnose this as being the problem? Should I be checking the temperature of the water in the header tank / that is overflowing?

I've tried to supply as much background info as possible, but I understand that I need to investigate further. If you have any pointers as to what I should check, please let me know.

  • Is the osso tank vented or is it a sealed tank?
    – d.george
    May 23, 2017 at 14:31
  • Unvented (need to type more characters)
    – Fat Monk
    May 23, 2017 at 14:33

2 Answers 2


Let me say first that I have never worked with this type of boiler, but the question you ask is similar to a few problems I have had with boiler systems here in the USA. I do not know if this is the same as your installation but I will tell you what I think may be the problem.

Someone specifies a larger pump than necessary or a larger replacement pump for an existing installation. The system is filled with water, vented, and put into operation. When the pump starts, the water pressure on the inlet side of the pump falls and the PRV (heating system pressure regulating valve) which is installed on the piping near the inlet side of the pump senses the drop in pressure and adds water to the system to correct for the reduced pressure.

When the boiler and the pump shut down, that extra head the larger pump exerted on the supply piping is the felt on the whole system. When this happens, the relief valve (which you call the overflow tank pipe) opens to get rid of the excess water that was introduced when the pump started. That extra water raises the whole system pressure above the set point of the "relief valve" when the system shuts down.

I do not know if this is what is causing your problem but it sounds similar. You may need to go back to the old pump size to correct the problem. Changing the pump just because a radiator was added or changed is a foolish mistake done by the uneducated in their trade.

  • I don't think there is an overflow 'valve' in the system, simply an overflow pipe that exits the header tank above the normal fill level as cut off by the ballcock. The pressure increase sounds like it makes sense, though, and my understanding of the system and what I've read so far were leading me in that kind of direction. What I'm missing is the detail of what could be causing this so your answer has given me some pointers to look at and think about.
    – Fat Monk
    May 23, 2017 at 13:48
  • Is the volume flow rate of the pump adjustable or is it set? May 23, 2017 at 16:07

I would adjust the ballcock in the header tank to reduce the level of water there.

When the boiler shuts off, it probably keeps the pump running for a short while to keep water circulating through the boiler to take away the residual heat. The zone valves are shut. Excess water is vented into the header tank raising the water level up above the overflow pipe. You need to provide space for that extra water by preventing the ballcock overfilling the header tank.

It might be that an over-specified pump is the root cause, but you can probably make compensating adjustments in the header tank.

You might also have a faulty ballcock valve that is not fully shutting off, contributing to the overfilling. They are relatively inexpensive and easy to replace. You can also replace worn parts in the valve.

enter image description here

  • Any idea why it would have just started doing this after the last drain down? Is it possible that it might resolve once the system settles down and doesn't need bleeding any more (I'm guessing there is still air in the system which could (?) result in a pressure differential - it usually takes a week or so of bleeding to clear of air after a drain down!)
    – Fat Monk
    May 23, 2017 at 13:49
  • Oh, I just added to the question that we only have the one small water tank in the loft - the water tank is fed from the water main (with a PRV inline, I believe).
    – Fat Monk
    May 23, 2017 at 13:57
  • If the picture showing the system layout, by RedGrittyBrick is a depiction of your system, then I have no idea what would cause your problem. Here in the states I have not seen this type of system installed in 80-90 years. A tank, with an overflow pipe, mounted in a high point in the house fell out of favor and was replaced with a sealed expansion tank mounted near the heating unit.
    – d.george
    May 24, 2017 at 9:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.