Suspended floor insulation choice

[osprey]

I have a suspended wooden floor, 1920s house, with 100mm joists, and about 250mm gap below joists. Sleeper walls are ventilated.

I have read MANY approaches to insulating these, and am bouncing between the choice of materials. PIR would give the best u value, but difficult to work with, needs space for battens. 'Loft roll' will settle over time, needs 120mm for spec U value etc...

So I was thinking of going the Rockool/Knauf 'loft roll' approach, with breathable membrane slung across the joists. But I was wondering if I could, instead, use the breathable membrane to support the PIR sheets. So, same 'hammock' approach as with 'loft roll', but no need to allow space for supporting battens and the membrane would reduce heat loss due to any gaps around PIR. PIR would be cut for close fit, and should slide along membrane better the rough wood joists.

Anyone see a major problem with this approach?

[ProDave]

A nylon netting I think called "netlon" or similar was always the normal way.  You lay it over the joists draping down, and fix it to the sides of each joist at the bottom to keep it taught across the bottom.  They don't need to be strong or thick, I ripped left over OSB sheets into thin strips for this.

 

Then lay the insulation in.  I like the Knauf frametherm type insulation, much nicer to handle than most glass wool products and it won't slump.  Cut it to your joist spacing with a panel saw while still on the roll still wrapped up.

 

Random link. Don't assume this is the best or cheapest supplier.

 

https://www.insulationsuperstore.co.uk/product/insulation-support-netting-2m-x-100m-white.html

[Redbeard]

Yes, you can definitely use breathable membrane as the 'hammock' and it arguably performs a 'belt-and-braces' function as an 'anti-wind-wash layer'. If you have nothing better to do for the next few hours read these. The BEIS Best practice document is interesting insofar as it is one of a series (IWI, solid floors, suspended floors and rooms-in-the-roof) which were published in (I think) 2022 but (as far as I am aware) barely publicised at all. I may not agree with absolutely everything in them but they have clearly used knowledgeable sources. Seach 'BEIS Best Practice' for the whole bunch.

 

In terms of material it is an expensive option, and it drops bits on you, but in terms of (semi-) rigidity flexible wood-fibre is soooo good. No dropping-out or drooping. Cut 20-30mm oversize, squeeze like a concertina and pop it in. I do 'support' it, as I want a membrane, but I am certain it would stay more-or-less forever without additional support.

 

Under-floor insulation guidance:

 

https://www.gov.uk/government/publications/insulating-suspended-timber-floors-best-practice

 

https://www.greenbuildingstore.co.uk/information-hub/insulating-suspended-timber-floors/

 

https://www.ecologicalbuildingsystems.com/post/best-practice-approach-insulating-suspended-timber-floors

 

http://www.sofiepelsmakers.com/suspended-timber-ground-floors.html

 

 

[saveasteading]

1 hour ago, ProDave said:

Knauf frametherm type insulation,

Me too. It can be compressed to sit snugly between joists with no gaps. Pir is a pain to cut, and to fit   and will inevitably have some gaps.

Then some insulating board, however thin,  over the top before boarding.

Edited April 10, 2023 by saveasteading

[osprey]

Thanks for the replies, but they are not really (mostly) answering my question, which was using PIR with the membrane hammock. Would the membrane mitigate some of the disadvantages of PIR? Knauf and similar have a worse U value than PIR, and PIR sheets come with a vapour control layer on one side. (The latter would be handy for a few areas which I can only get to from below).

 

(And thanks, Redbeard, I had already read those links you provided)

Edited April 10, 2023 by osprey

[nod]

The short answer is Yes it would 

[Sparrowhawk]

12 minutes ago, osprey said:

Thanks for the replies, but they are not really (mostly) answering my question, which was using PIR with the membrane hammock. Would the membrane mitigate some of the disadvantages of PIR?

Yes, an airtightness membrane will mitigate some of the disadavantages of PIR.

 

There's two components to well-functioning suspended floor insulation: airtightness and insulation. PIR is good at insulation, but the normal problem is getting it airtight when installing it because of the fit at the sides of the insulation. So if your membrane stops cold external air from being forced up the sides of the PIR and cooling the floor or insulation, it'll alleviate much of that problem.

 

Edit: But you will still get thermal bypass and your insulation will be less effective than you expect, because the air on both sides of the membrane will be at the same temperature. What you're avoiding is windwashing and larg-scale air circulation.

Edited April 10, 2023 by Sparrowhawk

[Iceverge]

 

Look at the above picture from the ecological building systems website. It's a good system in my opinion. 

 

The thermal resistance of a material is mute until you make sure you can install it properly. The above system makes sure no wind can blow through or around the insulation. 

 

PIR will perform worse than the cheapest mineral wool if it's not installed well and the mineral wool is. 

 

 

I would put. 

 

Breather membrane draped between joists. Stapled square  to the bottom of the joists with all joints taped and returned to the external walls and sealed here too. 

 

Good performance mineral wool fitted between the joists. 

 

T*G chipboard glued and screwed over the top. 

 

If you have space put a layer of PIR over the top of the joists under the chipboard. This will need to be between battens and all joints foamed and taped. 

 

 

 

 

 

 

Edited April 10, 2023 by Iceverge

[osprey]

12 minutes ago, Sparrowhawk said:

Yes, an airtightness membrane will mitigate some of the disadavantages of PIR

 

Edit: But you will still get thermal bypass and your insulation will be less effective than you expect, because the air on both sides of the membrane will be at the same temperature. What you're avoiding is windwashing and larg-scale air circulation.

If, in your edit, you mean thermal bypass due to poor fitting of the PIR, I would contend 1) that the gap would be intermittent  and small, and 2) due to the membrane, the air in the gap would be static. Air is a good insulator, as long as there is not movement due to convection currents, and, as heat rises and the room is warm, there would be negligible heat loss due to convection.

[osprey]

Iceverge, yes I have seen that. My suggestion would be as wind-proof as that, using the same membrane approach. Just with PIR instead of the mineral wool.

 

Incidently, wonder if, with 100mm joists, it would be worth letting the membrane go 20mm below the joist (while still stapling as shown) and fill to joist top.. This would allow mineral wool to achieve  U value to closer to 100mm PIR.

[Iceverge]

40 minutes ago, osprey said:

Iceverge, yes I have seen that. My suggestion would be as wind-proof as that, using the same membrane approach. Just with PIR instead of the mineral wool.

 

 

Mineral wool and other batt products is that they will fluff out against the sides and each other, really reducing any convection within the insulation layer.  Similarly the timber will expand and contract with changes in humidity and temperature and a flexible material will accommodate. 

 

With PIR, on the other hand, if you have a 5mm gap at install this will remain for all of time so I would make sure it was foamed with a permanently elastic foam in any case.

 

45 minutes ago, osprey said:

Incidently, wonder if, with 100mm joists, it would be worth letting the membrane go 20mm below the joist (while still stapling as shown) and fill to joist top.. This would allow mineral wool to achieve  U value to closer to 100mm PIR.

 

Yes no issues with this so long as you still have a clear ventilation path below of say 50-100mm. 

 

If you wanted to do a small bit of woodwork you could go as deeper with a few rips of 9mm OSB tacked onto the sides of the joists to keep it all in place. 

 

 

[Temp]

I think the membrane approach is best. 

 

If it had to be PIR I wondered if you could shoot long nails right through joists and into the side of a PIR block to holding up. Then foam any gaps. Might need a slightly different solution for the edges of the room.

 

I can think of a few ways to fix battens under existing joists without needing to fix up from below. 

[osprey]

Temp: both approaches use membrane for the support of insulation.

 

Iceverge: I would hope to do better than 5mm gap, but even with that, a static column of air with any warming coming from the top would be a pretty good insulator.

[Sparrowhawk]

3 hours ago, osprey said:

If, in your edit, you mean thermal bypass due to poor fitting of the PIR,

I do.

 

3 hours ago, osprey said:

I would contend 1) that the gap would be intermittent  and small,

Then your PIR cutting is better than mine, and any I've seen pre-installed in our house. It's keeping the blade perpendicular to the surface when cutting the PIR that's challenging - as a result there are voids. Gapotape negates much of this problem but is an extra expense.

 

3 hours ago, osprey said:

and 2) due to the membrane, the air in the gap would be static. Air is a good insulator, as long as there is not movement due to convection currents, and, as heat rises and the room is warm, there would be negligible heat loss due to convection.

True, the lack of air movement is going to make a difference.

 

Have you read https://aecb.net/thermal-bypass-risks-a-passivhaus-report-by-mark-siddall/ ?

Edited April 10, 2023 by Sparrowhawk

[saveasteading]

1 hour ago, osprey said:

would hope to do better than 5mm gap,

Hope is good!

But joists will be warped in length and vertically, so every cut will be different and approximate. The thicker the pir the more difficult to fit accurately.

Also, many talk of pir shrinking in time.

 

You will get a better theoretical rating with pir, but perhaps better insulation with more pliable materials. 

[osprey]

Sparrowhawk:

Yes I have read the paper (well, skimmed it, to be honest) thanks.

Most of it is concerned with walls (verticle surfaces, with a warmth from one side, rather than above, as with the present case), loft/ceilings (warmth from below) and whole house considerations. I do not think there is much of relevance to a 100mm verticle gap, warmth from above, enclosed by breathable membrane below and VCL above. 

Please correct me if I am wrong!

[markocosic]

If retaining the suspended floors I would:

 

- staple a "hammock" made of roofing membrane between the joists

- pop a sheet of EPS in the bottom of the hammock as a "base" to spread it - but make this base ~150 mm BELOW the bottom of the joists

- infill with mineral wool UNDERNEATH the joists

- infill with mineral wool BETWEEN the joists

- bubble glued chipboard over the top as the airtight layer/vapour retarder to prevent bulk movement to/from the house; taped to the walls and plastered over at the edges

- floor cover over that

 

This is similar to the ecological building systems photo above. Less resistant to wind-washing; but that's not a big issue on suspended floors. More than makes up for it by being able to double your thickness of mineral wool AND putting the whole of your joists on the warm side of the insulation.

 

The hammocks end up looking a little like like this - there's a lump of wood in the insulation fill which is the joist. Will ask matey if he has any photos of a floor done this way.

 

|XXXX___||XXXX___|

|XXXXXX||XXXXXX|

 

[Temp]

[Mod note: this reply and the next are responses to a post by a spammer that has now been deleted. Leaving the responses here in case of interest]

 

Kingspan and Celotex are companies. Both make a range of insulation boards both PIR and Phenolic with similar thermal conductivity.

 

Just two examples..

 

Kingspan - Kooltherm K108 - Phenolic - thermal conductivity of 0.019 W/mK

Celotex - GA4000 - PIR - thermal conductivity 0.022 W/mK

 

The smaller the number the better.

 

 

 

 

[saveasteading]

Yes, best to avoid trade names, and there are more...quinntherm  xtratherm, recticel....

Asking for pir allows suppliers to offer whatever is the best value.