Wiring underfloor heating with 3 manifolds

Hi all, taken on a new build project and they have quite a large property with three separate manifolds. I have wired before with one manifold but getting slightly confused.

Do I need to take a fused spur to each individual manifold to power the pump or can this be done via the three cores back to the wiring centre.

I usually run 3 three cores back from the manifold to allow for the Live- Neutral and Earth. Boiler return and the feed to the pump.

Am i right in doing this? More concerned the 3 pumps will overload the boiler fused spur which is the traditional way we wire underfloor heating it is fed via the boiler spur.

Client has also said they want to use nest, I was planning on running 3 core to each room/zone but I’m googling and seem to be wireless? Does this connection work wirelessly when you say have 20 zones across the house?

Any advice id be grateful!

Surely each manifold needs a wiring centre/ controller for the valves and pump and links to the thermostats, eg Heatmiser UF8 v2 RF. You can use a local source of power for each manifold controller, then either rf or volt free relay contacts back to boiler
I'm not sure Nest is cut out to do this - my only experience is with Heatmiser, Vaillant and Honeywell controls, and I would go Heatmiser and NeoAir thermostats for what you're describing.
Have you tried PlumbersForum?
PS do you know how many ufh loops on each manifold, and how many rooms, which will govern how many channels on the controllers?

Avo Mk8 said:

Surely each manifold needs a wiring centre/ controller for the valves and pump and links to the thermostats, eg Heatmiser UF8 v2 RF. You can use a local source of power for each manifold controller, then either rf or volt free relay contacts back to boiler
I'm not sure Nest is cut out to do this - my only experience is with Heatmiser, Vaillant and Honeywell controls, and I would go Heatmiser and NeoAir thermostats for what you're describing.
Have you tried PlumbersForum?
PS do you know how many ufh loops on each manifold, and how many rooms, which will govern how many channels on the controllers?

Click to expand...

Yes each manifold will have a controller, have experience with heatmiser and did recommend it. I guess each manifold will just have its own valve back at the boiler and will be wired individually. I will give each it’s own fused spur to be covered.

Seanlar said:

Yes each manifold will have a controller, have experience with heatmiser and did recommend it. I guess each manifold will just have its own valve back at the boiler and will be wired individually. I will give each it’s own fused spur to be covered.

Click to expand...

Are you envisaging each manifold being one of these (with a pump and blending valve If necessary?)

Avo Mk8 said:

Are you envisaging each manifold being one of these (with a pump and blending valve If necessary?)
View attachment 112354

Click to expand...

Yes exactly. Though house is set over 3 floors with a manifold on each floor. Valves will be in the plant room in the garage so will provide a spur to each manifold and wire back to plant room to allow for boiler return to open the valve.

Seanlar said:

I will give each it’s own fused spur to be covered.

Click to expand...

Then you will need to make sure that each is kept electrically separate by using relays or a proprietary control/wiring centre with volt free outputs for the signal to the boiler.

Personally I think giving each manifold a separate feed is overcomplicating things. A single feed for the whole heating system with a 3 pole isolator at each manifold would be my preference.

Seanlar said:

Am i right in doing this? More concerned the 3 pumps will overload the boiler fused spur

Click to expand...

Theres no need to be concerned, just calculate the total power consumption of everything connected. Each pump will be around 60W so you could comfortably have 10 on the 3A fused supply.

Seanlar said:

Client has also said they want to use nest, I was planning on running 3 core to each room/zone but I’m googling and seem to be wireless?

Click to expand...

The nest thermostat can be connected either wirelessly or wired by a 12V connection to the base unit, which is known as a heatlink. Each heat link is a simple L,N in and a volt free output.
The terminals themselves are quite small so it is usually easier to use a core for L and another core to the common of the output rather than trying to link the two together within the heat link itself.

Can't quote number without going to office to look it up ...
But regs require the system to have a single point of isolation. Makes sense as you don't want to be working on (e.g.) the boiler, knock off it's isolator, then get a belt from a control signal coming from another feed. But that would not prevent you having a higher rating feed and fuse it down as required. The Heatmiser wiring centers mentioned above have an internal fuse.
If using 2-port zone valves, there is isolation between the motor and switch. It would be possible to have separate feeds to each wiring center and power the related zone valve from that - then parallel the switches to fire the boiler. I would suggest it would be a poor design IMO.

I would suggest you need a 4 core from central wiring centre to each UFH wiring centre - L, N, E out and return (from "Boiler" relay). A 5 core would give flexibility - e.g. using the Boiler relay in the UFH wiring centers as volt free contacts.
Then standard S-plan wiring centrally.

For each zone, future proofing neans a 3 core (+E) from wiring centre to room controller/stat - L, N, switched live.
If using wireless, you just wire the receivers the same locally to the wiring center.

Simon47 said:

But regs require the system to have a single point of isolation.

Click to expand...

Not necessarily.
I haven't got the BBB to hand, but Part 9 (9.1 (e)) of NICEIC EIC implies more than one device can be employed.

Just to add another angle on this:
I have two UFH manifolds with pumps, a Heatmiser UF8 controller on each, a Vaillant heat pump (outside!) and its control units inside.
The Vaillant control stuff is in one place in the house, the downstairs manifold in another, and the upstairs manifold in another place.
Each of those three things has a separate FCU, not necessarily on the same CU circuit, and that is OK because there are no mains interconnects between these three areas, and I don't forsee any time when there will need to be. The only circuit that is common to all three is a low voltage (isolated) signal cable using volt free contacts in the UF8's to signal 'call for heat'.
I believe the system (not installed by me) is fully compliant with regs.
I don't see why, in the above example, there would be any need to run a common supply to all three places in the house. If maintenance needs to be performed on one area, it is only necessary to turn off the local FCU, and that is then safe to work on.

PS I should add that our room stats are RF. I guess if they were wired mains there might be more of an argument for a single point of isolation!

I think there's "shades of grey" involved.
I can't find the regulation I was thinking of, and it's possibly not specific to heating systems. But at one extreme you could have a mains powered thermostat (or control system) providing a control signal to a boiler - while the boiler is off a different supply. In that case, you could knock off the isolation next to the boiler - and get a zap from a supply from elsewhere.
At the other extreme, where there's a radio link, then no risk of that.
In between we have the sort of situation being discussed here - using the volt-free relays in a UFH wiring center, or the volt-free switch in a zone valve to switch something on a different supply. I would say that's OK given two things : 1) everywhere with multiple supplies is properly labelled (537.1.2), and 2) anyone working on the system is capable of understanding what the label means.
Now, it's something I've done in a commercial environment ("many" heat/cool fan-coil units, relay controlled with local supply and centralised timer, relay isolated demand signal to boilers and chiller plants) - it would have been completely impractical to do otherwise. In a domestic environment I'd be "cautious" - lets face it, some elements of some trades have a reputation for struggling to cope with S-plan wiring, do you really think they could cope with multiple supplies safely ?

Simon47 said:

In between we have the sort of situation being discussed here - using the volt-free relays in a UFH wiring center, or the volt-free switch in a zone valve to switch something on a different supply.

Click to expand...

I think that's the point I was trying to illustrate. If volt free contacts on a device on one supply are switching mains voltage coming from another supply, I agree that creates a hazard, and that's not acceptable.
But if the volt free contacts are switching a properly isolated electronic sensing circuit, eg SELV or equivalent, as we have on the input to electronic controls these days, then that hazard shouldn't exist.