Spiral or serpentine UFH layout - which is better?

[low_and_there]

Does anyone have any thoughts on whether the spiral design layout for UFH is significantly more performant than the serpentine one?
 

We’re planning on wet UFH throughout and I’ve had conflicting recommendations on layout (havent yet had actual design done).
 

I’ve been advised that the spiral design is better as it takes the heat first to the outer edges of the room, which will benefit from the heat more, but two other people I’ve spoken to (all people who have some experience or expertise) suggest it makes no odds, especially as we’ll have MVHR moving air around.
 

If spiral is much better, does anyone know somewhere we can get 22mm chipboard overlay boards (I can’t find anyone who supplies those with a spiral design - presumably it would need to be custom..) 

 

For context, we’ll have a very well insulated and airtight subfloor (0.13 u value) under the UFH, plan is to have it served by a ASHP.

 

thanks

[Olf]

Define 'better' and then you have the answer

Spiral: more uniform floor temperature, harder to plan laying but easier to execute (no 180deg bends except the very centre), and because of that lower flow resisitance

Serpentine: gradient of floor temperature as the water gets colder, not much thinking of the layout but fight with the ends to not to kink the pipe and stop it from raising.

The gradient is not necessarily bad thing, if the house requires low heat output (so what floor can give is not a limiting factor), having cold end of a loop running along the walls further reduces heat losses.

8 hours ago, low_and_there said:

If spiral is much better, does anyone know somewhere we can get 22mm chipboard overlay boards

And there is a 'better' of overlay boards: serpentine have constant spacing of linear channels, so you can have them ready made. Spiral requires constatly changing radius, so it would have to be custom

 

8 hours ago, low_and_there said:

as we’ll have MVHR moving air around

Generally MVHR moves less heat than UFH can produce, so don't count on that

[JohnMo]

If well insulated flow temps are so low you won't be able to tell the difference.  So choose whichever. If you are using a contractor to install go with there recommended layout.

 

MVHR does not move heat, as flow rates are way to low.  It's only replacing around a 1/3 the volume each hour.

 

What's your floor build up?

[low_and_there]

11 hours ago, JohnMo said:

If well insulated flow temps are so low you won't be able to tell the difference.  So choose whichever. If you are using a contractor to install go with there recommended layout.

Ok, thanks for the steer on this.  I’ll be laying the boards and pipes myself, plumber will do manifolds and commissioning (that’s the current plan anyway). Just now I’m gathering quotes.
 

 

 

[low_and_there]

11 hours ago, JohnMo said:

 

What's your floor build up?

Thinking is for the structural floor:

 

1) 22mm chipboard with UFH pipes

2) 6mm or 9mm ply

on top of joists.

 

Then finished floor will be either:

a) 16-18mm engineered wood or

b) Thin self levelling screed and 2mm Marmoleum (Lino)

depending on the room.

 

 Any thoughts on that?

[JohnMo]

Insulation?

[PeterW]

22 hours ago, low_and_there said:

especially as we’ll have MVHR moving air around.

 It doesn’t … flow rates are so low it’s not capable of huge heat transfer. 
 

1 hour ago, low_and_there said:

Thin self levelling screed and 2mm Marmoleum (Lino)

 

Is that over 6mm ply ..? That will crack and move. 
 

What’s between the pipes ..? And who’s designed the bulld up .?

[low_and_there]

13 hours ago, JohnMo said:

Insulation?

between 250 and 300mm of Cosywool between and under the joists.

Edited June 19, 2022 by low_and_there

[low_and_there]

12 hours ago, PeterW said:

What’s between the pipes ..?

I don’t get the question - the proposed approach is to lay the pipes in routed grooves in the Chipboard… so between the pipes - horizontally - it’s chipboard…?

[PeterW]

7 hours ago, low_and_there said:

I don’t get the question - the proposed approach is to lay the pipes in routed grooves in the Chipboard… so between the pipes - horizontally - it’s chipboard…?

That won’t work. Your floor will break as you will have routed out 2/3rd of its thickness and no amount of ply will make up for it. Why not fit spreader plates under the chipboard and then you’ve got a flat surface to start from. 

[low_and_there]

1 hour ago, PeterW said:

That won’t work. Your floor will break as you will have routed out 2/3rd of its thickness and no amount of ply will make up for it. Why not fit spreader plates under the chipboard and then you’ve got a flat surface to start from. 

It won’t be 2/3 or the chipboard that is routed out, although a non-trivial amount will be. But it’s not that much - 12mm out of 22mm across some of the board is probably less than 7% of its surface (at a guess). 
 

This is the exact build up offered by several UFH suppliers, for example: https://www.continal.co.uk/systems/suspended-floor

 

I’d be surprised if they were all pedalling something that will break…

 

 

[low_and_there]

1 hour ago, PeterW said:

Why not fit spreader plates under the chipboard

Our joists are only 4 inches deep so I’m disinclined to notch them.

 

I appreciate your critique - have you a suggested alternative floor build up for UFH based on overboarding?

 

thanks

[PeterW]

49 minutes ago, low_and_there said:

Our joists are only 4 inches deep so I’m disinclined to notch them.

Ok slightly lost now as when asked about insulation you said this :

 

9 hours ago, low_and_there said:

between 250 and 300mm of Cosywool between and under the joists.

 

So are you saying you only have 4” joists and the rest is an undercroft ..? How are you planning on suspending 200mm of fibre insulation ..?! You also don’t need to notch the pipes -

just drop under them at the ends and tack to the bottom of the joist with a pipe clip. 

 

54 minutes ago, low_and_there said:

won’t be 2/3 or the chipboard that is routed out, although a non-trivial amount will be. But it’s not that much - 12mm out of 22mm across some of the board is probably less than 7% of its surface (at a guess). 
 

This is the exact build up offered by several UFH suppliers

OK so unless you lay the pipes perpendicular to your joists, it creates a crack plane along the pipes. There is also a difference between routing out a few pipes and using an engineered product. This stuff is also foil lined in the grooves, only fits 12mm pipe and is eye wateringly expensive. It’s a last resort when nothing else will work for very low profile jobs.
 

To fit Marmoleum or any of the Lino type products you need a seriously rigid floor - that means the ply has to be glued and screwed to the joists, and then you need a decent compound over the top if you’re looking to level between floors. It cannot flex or move or if will crack and cause issues. 
 

I would look at this differently - what is your limiting factor on height you are working with, and how old is the building you’re working with ..? What heat output do you need, and what other insulation have you got in the building fabric as all of this is a contributing factor. Without all that information I can’t suggest anything other than using a spreader system although even that has its limits if you have high heat requirements. 
 

What the manufacturer of the boards that use 12mm pipe won’t have told you is the output is 20% lower per square metre, so you need either more pipe per metre or a higher temperature. That then limits your floor choices as engineered wood and Lino both have maximum flow temperatures. 
 

 

[low_and_there]

42 minutes ago, PeterW said:

How are you planning on suspending 200mm of fibre insulation

The void beneath is big so I’ve laid batons underneath, hung from joist hangers from the sleeper wall timber, and then a breather membrane to hold the insulation. In one room I get 200mm below the joists, the other 150, leaving in excess of 250mm ventilation underneath- in most areas it’s more like 500mm (house is on a hill)

 

[low_and_there]

 

47 minutes ago, PeterW said:

I would look at this differently - what is your limiting factor on height you are working with, and how old is the building you’re working with ..? What heat output do you need, and what other insulation have you got in the building fabric

 

ok, cool, thanks, so:

 

Floor height build up max 55mm from joist level

 

house built c1905

 

IWI on front (80mm of woodfibre probably), EWI on rear (225mm) - working assumptions

Likely a 5KW Heat Pump

 

 

[PeterW]

1905 is solid 9” wall even with insulation (why wood fibre ..??) , so you’re needing in excess of 150W/m2 which you won’t get from an overlay system so you are better off on radiators. 5kW heat pump will be running 24/7 in winter unless you’ve sealed every hole in the building and like 15-16°C room temperatures. 
 

Have you done a proper heat loss calculation for the building and worked out the real numbers ..? 

[low_and_there]

11 minutes ago, PeterW said:

1905 is solid 9” wall even with insulation (why wood fibre ..??) , so you’re needing in excess of 150W/m2 which you won’t get from an overlay system so you are better off on radiators. 5kW heat pump will be running 24/7 in winter unless you’ve sealed every hole in the building and like 15-16°C room temperatures. 
 

Have you done a proper heat loss calculation for the building and worked out the real numbers ..? 

 

You seem very sure of your assertions - why do you say that an overlay UFH system will deliver less heat than rads in this context? Can you articulate your reasons?

 

Yes, I've done heat loss calcs.

 

Why woodfibre? Because it is a relatively sustainable form of insulation and suitable for solid brick as it's hygroscopic. I have stayed in houses that match this make up of build and insulation and they're not 16 degrees in winter - they're toasty and cosy.

 

[PeterW]

8 minutes ago, low_and_there said:

You seem very sure of your assertions - why do you say that an overlay UFH system will deliver less heat than rads in this context?

yes - an overlay system at delta 15°c can deliver a maximum of 65W/m2 using spacing of 150mm which they are fixed at. I can increase a rad to deliver using type 21/22 rads that can put around 4 times that into a room using delta 50°C if needed to reach the heat loss required. 
 

14 minutes ago, low_and_there said:

Yes, I've done heat loss calcs.

Using what method ..?

 

15 minutes ago, low_and_there said:

Because it is a relatively sustainable form of insulation and suitable for solid brick as it's hygroscopic

Hygroscopic insulation is irrelevant for a solid wall dependent on the outer final finish. A 9” solid brick wall isn’t breathable unless you use lime render outside and a full lime based solution inside - at that point you would be better using a VCL and 65mm PIR internally and increased ventilation - cheaper, quicker and easier. 

[low_and_there]

9 hours ago, PeterW said:

Hygroscopic insulation is irrelevant for a solid wall dependent on the outer final finish. A 9” solid brick wall isn’t breathable unless you use lime render outside and a full lime based solution inside - at that point you would be better using a VCL and 65mm PIR internally and increased ventilation - cheaper, quicker and easier.

It’s not irrelevant at all. If I have a preference for traditional materials that do less damage to the environment, that’s my preference. The outer wall is only rendered (in the original lime) at 1m and below and the rest brick, and internally I’m using lime. Because I want to and because the materials and labour cost isn’t the only driver: I have an interest in sustainability, where PIR falls short. Your opinion is your own, but please respect when others may have a different one.

[PeterW]

1 hour ago, low_and_there said:

If I have a preference for traditional materials that do less damage to the environment, that’s my preference.

I’m not questioning that however you’re using a product that has a low r-value and will need to put significantly more heat into the building through whatever source you choose to offset the low thermal performance. The eco credentials of the insulation become pointless if you need to massively increase your carbon footprint to heat the property ! The whole lifecycle cost of the products gets ignored and the sustainability argument goes out of the window - we have regularly had this argument with clients who want net zero properties, and we have to get into the lifecycle cost of materials which is probably more important than the baseline initial “environmental” cost of the product itself.
 

If you have a limited insulation thickness on the front wall then consider a better performing product to get at least a BRegs minimum score for the element and then use the wood fibre elsewhere. 
 

 

[low_and_there]

Thanks for articulating the rationale behind your earlier post, @PeterW, ISWYM more clearly - appreciate it.

 

Back to the UFH, I am now looking at a 'pug' system with dry screed - doesn't weigh as much as the wet (I've read 25kg/m2), so I think the joists can take the weight (when I'd looked at wet screed earlier, I'd read that it is super heavy and not recommended over joists as it'll crack). 

 

Do you have a view on the Pug option? It looks like it can deliver a higher w/m2 than chipboard (90 is claimed by Wunda) and leads to a lower build up than wet (I'd read 65mm for wet but looks like dry can be 30mm).

 

For example: https://www.wundagroup.com/wp-content/uploads/2019/10/F05-Wunda-Pug-Screed-fact-sheet-30-4-2018.pdf

 

Thanks

 

 

 

[PeterW]

Pug mix works nicely and you can strengthen the floor using 18mm ply/chipboard and then cross batten with standard roof battens at 300mm centres, glued and screwed to the sub floor. This leaves you with space for 100mm spacing for standard 16mm pipework. A semi-dry pug mix over the top then capped with VCL and 12mm ply screwed to the battens will give you a 55mm build up. If you suspend the insulation from below using netting or even cheap landscape fabric it will make it easier to install. 

[Nickfromwales]

33 minutes ago, PeterW said:

Pug mix works nicely and you can strengthen the floor using 18mm ply/chipboard and then cross batten with standard roof battens at 300mm centres, glued and screwed to the sub floor. This leaves you with space for 100mm spacing for standard 16mm pipework. A semi-dry pug mix over the top then capped with VCL and 12mm ply screwed to the battens will give you a 55mm build up. If you suspend the insulation from below using netting or even cheap landscape fabric it will make it easier to install. 

Yup. The caveat is that using up the depth of the floor joist means a diminishing space for a high-performance ( or indeed any other ) insulation under the pug screed.

Needs a bit more thought  Why not fill the depth of the joist space with the best possible insulation, atop the wool product, as a secondary higher ( highest ) performing insulation layer and THEN you can go for a 22mm P5 deck board minus the troughs. 6mm ply on that and you're sorted.

[PeterW]

1 hour ago, Nickfromwales said:

The caveat is that using up the depth of the floor joist means a diminishing space for a high-performance ( or indeed any other ) insulation under the pug screed.

Suspended floor my owd fruit so he’s got plenty of room for suspended insulation under the floor/joists etc 😁

[Nickfromwales]

16 hours ago, low_and_there said:

The void beneath is big so I’ve laid batons underneath, hung from joist hangers from the sleeper wall timber, and then a breather membrane to hold the insulation. In one room I get 200mm below the joists, the other 150, leaving in excess of 250mm ventilation underneath- in most areas it’s more like 500mm (house is on a hill)

 

A layer of PIR directly under spreader plates ( set out meticulously so as to make the plates very slightly distended eg for excellent surface to surface contact with the P5 ) will add to the mineral wool and pay huge dividends.

The wood fibre could be used instead, but if it were my house I’d use PIR for the uplift in the performance credentials.

 

21 minutes ago, PeterW said:

Suspended floor my owd fruit so he’s got plenty of room for suspended insulation under the floor/joists etc 😁

Yes, already noted, Sherlock 🤪.
 

Just saying the underside of the heated pug tray would be only protected by the wool and also subject to ventilation heat loss from the free air movement in those underfloor voids. 
Thus, the PIR, sealed and made draught-proof, would ( should ) be a minimum spec under the pug tray. Prob struggle to get 25mm under unless the tray bottom was 11mm ply and in-filled with PIR before loading with the pug mix. 
 

Wool > PIR > spreader plates c/w 3x16mm pipes per tray / void > 22mm P5 > 6mm ply > floor covering = 👌