Rads upstairs cold when underfloor heating on downstairs

[campbep]

Ah, right thanks.

I've only got 7-way manifolds on the UFH with five individual room thermostats. So, if the room 'stats don't demand heat from the system and if the TMV is faulty, where's all the excess heat going?

 

[SimonH2]

Ah, right thanks.

I've only got 7-way manifolds on the UFH with five individual room thermostats. So, if the room 'stats don't demand heat from the system and if the TMV is faulty, where's all the excess heat going?

If you turn all the stats down, do the upstairs rads get warm ? Don't forget that the actuator heads may take as long as 20 minutes to close the valves - I'm assuming there are thermal actuators on the UFH zone valves.

But by far the quickest and easiest thing to check is the boiler flow temperature. If it's a "boiler not hot enough" problem then the pipes won;t be all that hot to touch. If it's a "not enough flow" problem then the boiler flow pipe will be untouchably hot.

At the UFH manifold, the difference between boiler flow and UFH loop temperatures should be very easy to tell.

 

[campbep]

Of the 2 pipes coming out of the top of the boiler, one is too hot to touch and the other is hot.

At the UFH manifold, the pipe between the electric valve and stat is warm (boiler flow one?) and the other pipe is pretty cold (loop?)

Have turned all UFH room 'stats down and will report back with result

 

[campbep]

Turning all UFH ststs down has not made a real difference to upstairs rad temps

 

[SimonH2]

Of the 2 pipes coming out of the top of the boiler, one is too hot to touch and the other is hot.

At the UFH manifold, the pipe between the electric valve and stat is warm (boiler flow one?) and the other pipe is pretty cold (loop?)

Sounds like a flow rate problem then - either a blockage, or more likely an issue with the pump. Perhaps one of the more experienced people would care to suggest what is most likely - are impeller failures common ?

One other check (before starting to take things apart or replace bits) comes to mind that would confirm it. You need the system cold (or at least cool), and then get an assistant to turn on the stat/timer for upstairs (with downstairs turned off). You are looking to see how fast the hot pipe heats up, and more importantly, how fast the heat travels along the pipe. If it's pump failure, then the flow pipe will heat up close to the boiler, but the heat will travel fairly slowly along the pipes - I suspect your upstairs heating has only been running by thermosyphon where the less dense (lighter) hot water rises up one pipe, and the denser (heavier) cold water drops down the other, all by gravity.

 

[campbep]

Ok

Did a cold start with just the upstairs rads. One of the pipes at the top of the boiler went from luke warm to cold very quickly and the other one got very hot quite quickly. The pipes upstairs around the pump and valve got hot within a minute or so of the boiler firing up.

A couple of the rads got hot reasonably quickly but some of them took a long time to get warm and a couple stayed cold.

 

[SimonH2]

Did a cold start with just the upstairs rads. One of the pipes at the top of the boiler went from luke warm to cold very quickly and the other one got very hot quite quickly. The pipes upstairs around the pump and valve got hot within a minute or so of the boiler firing up.

A couple of the rads got hot reasonably quickly but some of them took a long time to get warm and a couple stayed cold.

I think that confirms that it's a lack of flow problem - with the return pipe cold AND a good flow rate, the boiler shouldn't have been able to get the outlet hot all that quickly. Also, the pipes upstairs should have got hot quite quickly - tracking the rise in flow temp from the boiler.

I can't see it being a blockage since the system has been working for years. So either the pump is faulty, or the motorised valve isn't opening fully. The head should come off the valve, and you can manually turn the spindle - the flat tang is parallel to the hole or butterfly, so if it's parallel to the pipe then it's fully open.

After that, it's time to start taking bits apart unless someone has a better idea.

 

[campbep]

Thanks for the info. Turned out that the pump was compromised and there is some muck in the system. Also the expansion vessel had lost pressure and needed replacing.

 

[SimonH2]

Nice one. It's good to hear when someone has solved the problem.