Extension/Original floor UFH thoughts

[osprey]

I am getting close to starting my extension build, and keep second-guessing myself on heating method, so would appreciate thoughts on my current plan.

 

The extension will be a 5 x 5m single-height extension to my 1920s built house. It will be knocked through to the  original  kitchen, to make a kitchen/diner/utility room. The new kitchen part will be 5 x 3m. The floors will be a mix of new insulated slab over block/beam in the extension/dinner and suspended wooden floor, with a smallish area (3 x 1m) area of old solid floor, in the kitchen area.

 

The extension will be built to latest insulation standards, and the suspended floor has 100mm joists, so 100mm celotex insulation will be relatively easy to install. The kitchen area has 50mm filled cavity, and will have another 50mm insulated plasterboard added internally. The current floor has ceramic tiles, cement screed, and plywood on top of the original planks, so it has 30-35mm of height build already, that I can remove and install UFH.

 

I would like to install a retrofit UFH system, something like the Wunda system, across all three floor types, divided into 2 heating areas (kitchen and diner). 

 

I would also like to do the hallway (a third heating area) in the old part of the house with the same system, as it currently has the same floor covering over suspended wooden floor, which has to be removed wiring reasons anyway. I would leave the existing radiators in place here, as well as in the rest of the ground floor (living and sturdy rooms) and upstairs, which are also in the old part of the house.

 

Proposed floor covering over the pipes would be 10mm latex self levelling compound and Amtico.

 

Any thoughts of viability of such a system/layout?

Thanks.

 

[Nickfromwales]

I doubt 10mm latex over an overlay system would not give up eventually. Prob need the thicker alu faced system, I think I said elsewhere its the 400micron panel, and set 6mm cement / hardie boards on to those with tile adhesive. You'd then put feathering compound over those and then the LVT.

42 minutes ago, osprey said:

The new kitchen part will be 5 x 3m

I'd not use the overlay in there, and instead look to set the UFH pipes into the slab /screed. The issue with the change in floor types and UFH discipliners will be the heat up / cool down times (responsiveness), but not life or death. It would just require wall stats in the different zones / areas to allow each to function independently and run over / under dependant on the time and temps.

 

You'll also need to design it to have a single flow temp, so tbh I'd dig out the 3x1m section and fill that in to match the rest of the area so they 'play nicely'.

[osprey]

Correction to original post: a large part of the kitchen floor will be covered by cupboard units, of course, so the 'solid' part under UFH would only be 1 x 1.8m. The remaining 'suspended' part under UFH would be more like 12sq metres.

 

(The diner is new solid floor, kitchen is  mostly suspended wood, if that was not clear in OP)

Edited February 21 by osprey

[osprey]

 

19 hours ago, Nickfromwales said:

 

I'd not use the overlay in there, and instead look to set the UFH pipes into the slab /screed. The issue with the change in floor types and UFH discipliners will be the heat up / cool down times (responsiveness), but not life or death. It would just require wall stats in the different zones / areas to allow each to function independently and run over / under dependant on the time and temps.

 

You'll also need to design it to have a single flow temp, so tbh I'd dig out the 3x1m section and fill that in to match the rest of the area so they 'play nicely'.

Using the in-screed system in the new extension was my first thought, but:

1) Amtico flooring must not be heated above 27⁰, so Wunda supply a temperature sensor for the floor. Not sure if this would work for screed, but even if it did, the 'access' pipe lay to the extension would run with the surface mount stuff across the kitchen, and may get too hot

2) Having a mix of UFH and radiators  AND  a mix of floor types put me off having a mix of UFH systems as well!

 

[JohnMo]

I think I would make life easier for myself, nice radiator or two, maybe plinth heaters in the kitchen. Doing UFH well with multiple floor make ups isn't easy. Wunda etc. will sell you everything but doesn't make it a good or cost effective solution.

[Nickfromwales]

5 hours ago, JohnMo said:

I think I would make life easier for myself, nice radiator or two, maybe plinth heaters in the kitchen. Doing UFH well with multiple floor make ups isn't easy. Wunda etc. will sell you everything but doesn't make it a good or cost effective solution.

Yup, I have to agree as I think this will be an expensive disappointment. I certainly wouldn’t be looking at blowing money on both types of heating, with the UFH being a hybrid mix of solutions, plus this would be VERY difficult to control / balance too. 
 

Very good shout about the plinth heaters, as when the arse of winter kicks in you will really feel the benefit of that additional auxiliary heat source. Tall narrow ‘decorative’ or column type rads don’t give off a huge amount of heat via convection, so at the least these need to be as far towards the end of the rooms as possible to cast the heat to where it’s most needed.

 

I wouldn’t rule out putting electric UFH in here if there was a tiled floor, but with LVT you’ll struggle to do this; it would have to be set into the self levelling compound that the LVT needs to be installed over. Can be done and set to not get hotter than the 27^oC. If this is the forever home then consider that for comfort with the acceptance of the high running costs, but on the further understanding that it would only be used sporadically.

 

Filling the cavities with blown bonded beads and drought proofing will be very good measures to reduce the actual heat requirements, and would be another life long (and wise) investment. 
 

Use the money you just saved by not fitting Mary Shelley’s UFH system    

[osprey]

18 hours ago, Nickfromwales said:

Yup, I have to agree as I think this will be an expensive disappointment. I certainly wouldn’t be looking at blowing money on both types of heating, with the UFH being a hybrid mix of solutions, plus this would be VERY difficult to control / balance too. 
............

Use the money you just saved by not fitting Mary Shelley’s UFH system    

I think that is overstating it a bit.

 

Just to be clear, in the new kitchen diner it is ONE system over two floor types - insulated solid and insulated suspended (the uninsulated solid part is small enough to leave out).

 

 

[JohnMo]

2 hours ago, osprey said:

new kitchen diner it is ONE system over two floor types - insulated solid and insulated suspended

There is a thread from about year ago, someone did something similar and one part of the floor was great the other rubbish. The different heat profiles although small, had an impact on the system worked.

 

At very least you need to ensure each floor type has its own UFH loops, so the loops do not have to work in both areas. Then you can balance the flows to get the system to work correctly.

 

Do not try to mix radiators and UFH in the same area, you will have two systems with very different inertia fighting controls.

 

Your money, your house, opinions given as requested. KISS (Keep it simple stupid) is the best policy for heating systems, zones and mixed systems are normally worse, not better performing.

On 21/02/2025 at 19:50, osprey said:

thoughts of viability of such a system/layout?

 

[Nickfromwales]

5 hours ago, osprey said:

I think that is overstating it a bit.

 

Just to be clear, in the new kitchen diner it is ONE system over two floor types - insulated solid and insulated suspended (the uninsulated solid part is small enough to leave out).

Not really an overstatement, just physics says that the different downwards heat losses will give you two different results. I’m happy to comment on doing this vs not doing it, whichever you find most helpful. FWIW, I would have UFH in a heartbeat, even if it was less than great to use (running cost wise), but that’s me and I don’t assume people who come here want duff advice from any one person.
 

I am a fan of the Wunda stuff tbh, but their sales patter on the website oozes positivity and doesn’t really address the pitfalls of failing to prepare. I guess their designers would be the ones to give you more of an assurance, and more importantly offer an underwritten solution for heating output and controls, so I guess start there and see what their feedback is. When you factor in the SLC, I’d defo stay at 10mm or a bit more if possible. I ‘Wunda’ how good their support will be regarding overall methodology. 
 

The floor will get warm, that’s a given, it’s more about whether or not the spaces be sufficiently heated. The suspended area should be a cold ventilated space so this will need improving before laying the UFH over it, if possible, with some additional draught-proofing to prevent any airflow from the voids up into the rooms. This will pay huge dividends regardless of how you heat the spaces.
 

I think the overlay boards get bonded down, so if the joints are all foamed then this should deal with any residual infiltration. Foam all around the perimeter too, whilst the skirting boards are off, but I’m sure you’ll need to do this to create a moat for the SLC anyways; any gaps you leave for air to travel up is where all the SLC will end up disappearing down. That’s real fun to solve half way through a big pour!!
 

I’m not plucking my comments out of thin air btw, I’m just saying what I’ve seen (and had to rip out and redo) over 30+ years of installing heating and hot water systems in many different arenas . I’ve seen UFH done poorly more times than well, especially retrofit solutions or systems installed by “turnkey builders” with robot plumbers who just need to plough on to get a pay cheque every Friday. 

If it’s an overlay then you’d have to go over the original section anyway to keep the floor levels the same, so you may as well install the UFH there as well so the floor is at least warmed as the difference where it goes from heated to un-heated would be obvious and likely something you’d find intolerably cold when stepping on one or the other, even with LVT.

 

If going with the SLC option ( Wunda say 5-10mm  wish they’d be a bit more specific) I’d recommend using the Mapei construction screed with fibres vs the Ultra 2-part. I used both regularly and they’re both excellent, but I think the fibres would be a good idea where the system goes over the suspended floors.

 

As you’re installing LVT the effect between heated / unheated areas should be minimal, lesser again if you walk on it wearing socks, but this will be the worst covering choice for getting heat into the room from the water in the UFH pipes so you need to understand this fully before proceeding.
 

It’ll just be a lot of money and effort to do, but it can be done, so we’re just replying to you here to prepare you for the journey ahead if you do decide to go down this route. It’s better to discuss this to death here (for free), beforehand, than be coming back to find out how to resolve it and to be told “you can’t”.  

 

 

[osprey]

On 23/02/2025 at 12:57, Nickfromwales said:

Not really an overstatement, just physics says that the different downwards heat losses will give you two different results. I’m happy to comment on doing this vs not doing it, whichever you find most helpful. FWIW, I would have UFH in a heartbeat, even if it was less than great to use (running cost wise), but that’s me and I don’t assume people who come here want duff advice from any one person.
 

I am a fan of the Wunda stuff tbh, but their sales patter on the website oozes positivity and doesn’t really address the pitfalls of failing to prepare. I guess their designers would be the ones to give you more of an assurance, and more importantly offer an underwritten solution for heating output and controls, so I guess start there and see what their feedback is. When you factor in the SLC, I’d defo stay at 10mm or a bit more if possible. I ‘Wunda’ how good their support will be regarding overall methodology. 
 

The floor will get warm, that’s a given, it’s more about whether or not the spaces be sufficiently heated. The suspended area should be a cold ventilated space so this will need improving before laying the UFH over it, if possible, with some additional draught-proofing to prevent any airflow from the voids up into the rooms. This will pay huge dividends regardless of how you heat the spaces.
 

I think the overlay boards get bonded down, so if the joints are all foamed then this should deal with any residual infiltration. Foam all around the perimeter too, whilst the skirting boards are off, but I’m sure you’ll need to do this to create a moat for the SLC anyways; any gaps you leave for air to travel up is where all the SLC will end up disappearing down. That’s real fun to solve half way through a big pour!!
 

I’m not plucking my comments out of thin air btw, I’m just saying what I’ve seen (and had to rip out and redo) over 30+ years of installing heating and hot water systems in many different arenas . I’ve seen UFH done poorly more times than well, especially retrofit solutions or systems installed by “turnkey builders” with robot plumbers who just need to plough on to get a pay cheque every Friday. 

If it’s an overlay then you’d have to go over the original section anyway to keep the floor levels the same, so you may as well install the UFH there as well so the floor is at least warmed as the difference where it goes from heated to un-heated would be obvious and likely something you’d find intolerably cold when stepping on one or the other, even with LVT.

 

If going with the SLC option ( Wunda say 5-10mm  wish they’d be a bit more specific) I’d recommend using the Mapei construction screed with fibres vs the Ultra 2-part. I used both regularly and they’re both excellent, but I think the fibres would be a good idea where the system goes over the suspended floors.

 

As you’re installing LVT the effect between heated / unheated areas should be minimal, lesser again if you walk on it wearing socks, but this will be the worst covering choice for getting heat into the room from the water in the UFH pipes so you need to understand this fully before proceeding.
 

It’ll just be a lot of money and effort to do, but it can be done, so we’re just replying to you here to prepare you for the journey ahead if you do decide to go down this route. It’s better to discuss this to death here (for free), beforehand, than be coming back to find out how to resolve it and to be told “you can’t”.  

 

 

Great reply, thanks for the detailed help.

I am really worried about not having a warm room, after all  this work, but also worried about not taking the opportunity to have UFH where possible. (Even my architect has Wunda in and extension she did for her own similar age house). As I have already got 30mm height buildup, and I would want to insulate under the existing floors whatever the heating system anyway (100mm celotex), adding retrofit UFH seemed an obvious next step.

 

The plan is to have the 'new solid floor' diner part of a separate circuit to the 'suspended floor' kitchen, as different heat loss, as you say, but also cooking generates it's own heat...  Doing the hallway as well (while leaving radiator in place) was a more speculative idea - main heating by the radiator, but some radiating heat from the floor as background for now, but UFH heating in place should I get around to further insulating the rest of the house (e.g. with EWI). I appreciate balancing would be an issue, even with this relaxed requirement. But just the new kitchen/diner as UFH would be great, but a catastrophe if I had to rip it up to reinstall radiators!

 

With the screed, I am told it would take two layers to get the right depth. The builder could do the first layer, or the Amtico installer could do both. Not sure if the Amtico installer could/would use the Mapei stuff.

 

(WRT plinth heaters, is there no problem with hot air >>27⁰ being blown across the surface of Amtico?)

 

[Nickfromwales]

👍. 
 

Next step will be to contact the installer of the LVT and get their input, so you don’t accidentally get any of the process or prep wrong and they refuse to lay over it. 
 

LVT installers usually want a subfloor to lay their “feathering compound” on to, but they may be happy to do the whole pour. They’re the ones who will decide, ultimately, which SLC product gets used.

[osprey]

28 minutes ago, Nickfromwales said:

👍. 
 

Next step will be to contact the installer of the LVT and get their input, so you don’t accidentally get any of the process or prep wrong and they refuse to lay over it. 
 

LVT installers usually want a subfloor to lay their “feathering compound” on to, but they may be happy to do the whole pour. They’re the ones who will decide, ultimately, which SLC product gets used.

Yeah, when they quoted, the LV installers assumed they would be doing the 'double' pour, but they said they were happy if the UFH installers did the first one. (They checked, and said they use the 'same one as Wunda recommend', but I didn't know Wunda recommend two at that stage).