Could you make your diagram a little larger - my eyes are too old!


In general two pumps are used as the boiler circuit runs at a higher temperature than the UFH, so a secondary pump runs the (same) water around the UFH and the blending valve, just pulling a little of the hotter water from the boiler side as needed to keep the UFH at the right temperature. More like an adjustable autotransformer than galvanically isolated circuits, if you like.


If on the other hand you can arrange things so that the boiler runs at the desired UFH temperature, I don't see why it can't be done with a single pump.


   - Andy.

The maximum temperature this particular 4kW electric boiler will reach is 60 degrees, so with a bit adjustment it could be set to run at 55 degrees the design flow temperature of the UFH circuit, but presumably the system works better by blending the slightly hotter water from the boiler circuit with the return flow from the UFH.


There is an issue other than being told I have wired up too many pumps that I have not got clear in my head yet.


The actuators on the UFH don’t immediately respond to signals from the UFH controller, there is a 2-3 minute delay between getting the signal and being fully open, but the boiler “fires” immediately upon receiving the signal from the UFH controller, then overheats and locks out, this is made worse because the bypass has not been installed. I spoke to another plumber this morning and he questioned why the UFH controller fires the boiler before the actuators are fully open, but as far as I can see that is what supposed to happen with the bypass returning the water to the boiler and its thermostat controlling the boiler circuit flow temperature, but the boiler thermostat is on the boiler return and is not able to operate correctly without the water circulating.

The system was incomplete when I wired it, so I took the boiler pump out of its box and wired it up without it being plumbed in or being able to commission it.


Then I returned but the plumber wasn’t there, so I set the date and time on the pump display and tried adjusting the flow according to the instructions supplied, but eventually twigged it was the wrong way around, then the plumber went back to sort it out and said it doesn’t need the boiler pump at all. So rather than fitting a bypass in the boiler circuit he wants to take the boiler pump out.


I am convinced the UFH pump won’t circulate water through the boiler circuit on its own without the boiler pump.


The UFH pump is wired to the UFH controller and the boiler pump is wired to the boiler, so they each circuit runs separately at different temperatures.

The bypass could be a permanently connected radiator without a TRV, so it’s always hot when the system is in use.


But suggesting having a permanently connected radiator in a flat with UFH will not be a suggestion that goes down well.

I was a bit perplexed at first because I couldn't work out how the supply of hot water could be heated without a boiler pump. Then I noticed that you mentioned "electric boiler" so, presumably, hot water is heated by an immersion heater.


I don't think that any manifold should call for heat until at least one valve is fully open. If bog-standard S-plan can manage it, why not UFH?


I don't see any need for a bypass if this is done properly. If there is one, but no boiler pump, how would the water flow through it? ?

An Electric flow boiler.

Uses standard radiators, programmer, pump, room thermostat and thermostatic radiator valves. Models available for underfloor heating systems.


Well I never.


My understanding is that it  needs two pumps, one for the boiler circuit and one for the UFH circuits as they are two distinct circuits. 


Errrmmm, what is a boiler circuit?




Z.

The actuators on the UFH don’t immediately respond to signals from the UFH controller, there is a 2-3 minute delay between getting the signal and being fully open

Yup - mine do that. The UFH actuators aren't motors but little heating elements that warm up a wax capsule which in turn pushes the valve mechanism - rather like an ordinary TRV but extra heat opens the valve rather than closes it. I suppose they could be fitted with a microswitch to sense when the valve is fully open - but I've never seen one.


Normally (say with a gas boiler) it shouldn't be a problem as the boiler circuit will take that long (via the bypass) to get the water up to temerature anyway.


If it's an electric boiler that's only for the UFH, then the it would seem sensible to me to dispense with the blending value and just have a single circuit with one pump - a lot less to go wrong then. With a blending valve it might work with just one pump - I've sort of got that setup with mine, but it's fed from a thermal store rather than a boiler - but the pump is on the UFH side of the blending valve - when valve calls for heat there's enough "suction" to draw hot water into the UFH circuit from the thermal store.


  - Andy.