What's the best wall structure to balance performance and cost in a new stick frame extension, and is continuous insulation worth it?

[Jess Shannon]

We're building an extension on to an old stone cottage. The new foundation will be insulated raft with UFH. The extension will have a pitched roof section and a flat roof section with steel frame connecting to the old building.

A builder friend is helping us design it. He's suggested 220mm studs so we can have a thick layer of insulation. He said twin studs could end up costing more due to added labour over any material cost saving. But will the cold bridging mean it's not worth it in the long run? Should we apply an external layer of continuous insulation too?

I've mocked up his suggestions on ubakus to look at performance (edit it here).



And this is a build up that I've put together based on other research, with an external PIR sheathing (edit this one here). Would this kind of PIR board be able to wrap up and over the roof too?

[Mr Punter]

I think it may work well with either internal pir or insulated plasterboard, so move the 50mm celotex inside.  It follows the rule of thumb of having the most insulating and vapour impermeable materials inboard.

[SimonD]

 

36 minutes ago, Mr Punter said:

It follows the rule of thumb

 

Yes, I wouldn't have the pir on the outside either.

Principles are definitely to have insulation to reduce thermal bridging. In Sweden, for example, it's very common to have a continuous layer each side of the stud walls. You can then use smaller dimensioned stud. I wonder what the cost implications are of this compared to the cost of the 220 studs plus continuous layer?

[Redbeard]

16 minutes ago, SimonD said:

Yes, I wouldn't have the pir on the outside either.

 

... and I was just about to write that too. PIR externally amounts, in my view, to a VCL on the wrong side of the 'sandwich' - and I wince every time I see a TV show depicting it.

How about rigid wood-fibre with a scratch-coat of lime render before the battens and (?hit-and-miss?) larch?

[SimonD]

29 minutes ago, Redbeard said:

How about rigid wood-fibre

 

That's probably what I would go for too.

[sgt_woulds]

Using woodfibre & I-Joists with a direct rendered external sheathing board will give you a quick build, that will give you more decrement delay (probably more important than ultimate u-vale in our ever heating world.  You can also do away with membranes entirely and have no worries about moisture.  

 

 

 

The example build up uses (taped and sealed), 15mm OSB3 as a combined racking and V-VCL layer internally.

 

External, directly rendered woodfibre sheathing board.

 

I've shown 240mm I-joists as these are more common, but 220mm are available

 

I've also shown a service void behind the plasterboard as this makes wiring easier. 

 

 

 

 

Edited yesterday at 13:21 by sgt_woulds

[Crofter]

I designed my place with 50mm PIR on the inside, which allowed me to use 145mm studs whilst eliminating thermal bridging.

 

You want your wall build-up to be progressively more moisture permeable as you move outwards. So mine goes VCL, PIR, mineral wool, breather membrane.

[sgt_woulds]

If you used PIR, that is already acting as a VCL.  If you are careful with taping then an additional internal VCL would not be needed.

 

I have a combination of PIR and Woodfibre in part of my own roof (due to structural loading, woodfbre alone was too heavy) with the PIR Internal and taped as the VCL. 

[ADLIan]

2 hours ago, sgt_woulds said:

Using woodfibre & I-Joists with a direct rendered external sheathing board will give you a quick build, that will give you more decrement delay (probably more important than ultimate u-vale in our ever heating world.  You can also do away with membranes entirely and have no worries about moisture.

Do not render direct onto (any) insulation onto timber (or steel) frame construction!! It's not accepted good practice and lots of failures reported. About this time last year I was dealing with an architect looking at remedial works on this form of construction - sole plates and base of studs all rotted on 2 houses built approx 15 years ago.

 

Does PUR really form a VCL if the board joints are not sealed/taped in any way? Moisture vapour is a gas so will readily move through the joints. Discuss.......

[Mr Punter]

9 minutes ago, ADLIan said:

Do not render direct onto (any) insulation onto timber (or steel) frame construction!! It's not accepted good practice and lots of failures reported. About this time last year I was dealing with an architect looking at remedial works on this form of construction - sole plates and base of studs all rotted on 2 houses built approx 15 years ago.

 

Agreed.  Better is a render carrier board battened off the structure to leave minimum 25mm drained and vented cavity.

[JohnMo]

If doing a timber frame, look for cold bridging at the wooden sole plate and foundation also. Then design out.

[Iceverge]

I'll attract scorn here but I like a continuous layer of external insulation. It's keeps the studs and OSB sheathing warm and away from the elements. 

 

Provided you have adequate management of air movement through the structure (airtightness) and very good control over internal humidity levels the wall will quite happily dry to the inside provided you don't add any more low  permability layers like internal vapour barriors. 

 

Unfortunately airtightness is rarely done well and internal humidity control relies on the long term occupants diligence so there have been numerous building failures, hence the caution urged above. 

 

In your case the U value of the extension is lightly to be neither here nor there in the grand scheme of things. Aim for Bregs with something robust, easily buildable and cheap. 

 

What really will make it feel more comfortable is good air sealing. Make sure this is done well. I would avoid PIR internally as half of it will end up in the skip as offcuts and you'll be searching for studs to hit while screwing through it. Also it's virtually completely vapour closed so any accidental moisture in the wall will struggle to dry either in or out. Far easier to bang up a few 2*2s with a nail gun as a service cavity and mineral wool is fast and cheap to install. 

 

I really like blown cellulose for TF but it may not be worth getting the man for a small area of an extension. 

 

Importantly everything above is off the shelf at every builders merchant in the country. 

 

Robust, buildable. Cheap. 

 

 

 

 

 

Edited 17 hours ago by Iceverge

[Roger440]

7 hours ago, sgt_woulds said:

Using woodfibre & I-Joists with a direct rendered external sheathing board will give you a quick build, that will give you more decrement delay (probably more important than ultimate u-vale in our ever heating world.  You can also do away with membranes entirely and have no worries about moisture.  

 

 

 

The example build up uses (taped and sealed), 15mm OSB3 as a combined racking and V-VCL layer internally.

 

External, directly rendered woodfibre sheathing board.

 

I've shown 240mm I-joists as these are more common, but 220mm are available

 

I've also shown a service void behind the plasterboard as this makes wiring easier. 

 

 

 

 

 

As others have said, seems rather high risk and totally dependant on the render to keep it all dry.

 

Not for me that one.

[Roger440]

22 minutes ago, Iceverge said:

I'll attract scorn here but I like a continuous layer of external insulation. It's keeps the studs and OSB sheathing warm and away from the elements. 

 

Provided you have adequate management of air movement through the structure (airtightness) and very good control over internal humidity levels the wall will quite happily dry to the inside provided you don't add any more low  permability layers like internal vapour barriors. 

 

Unfortunately airtightness is rarely done well and internal humidity control relies on the long term occupants diligence so there have been numerous building failures, hence the caution urged above. 

 

In your case the U value of the extension is lightly to be neither here nor there in the grand scheme of things. Aim for Bregs with something robust, easily buildable and cheap. 

 

What really will make it feel more comfortable is good air sealing. Make sure this is done well. I would avoid PIR internally as half of it will end up in the skip as offcuts and you'll be searching for studs to hit while screwing through it. Also it's virtually completely vapour closed so any accidental moisture in the wall will struggle to dry either in or out. Far easier to bang up a few 2*2s with a nail gun as a service cavity and mineral wool is fast and cheap to install. 

 

I really like blown cellulose for TF but it may not be worth getting the man for a small area of an extension. 

 

Importantly everything above is off the shelf at every builders merchant in the country. 

 

Robust, buildable. Cheap. 

 

 

 

 

 

Im assuming that you would put some sort of cladding on battening on the outside of that build up?

 

[Crofter]

40 minutes ago, Iceverge said:

I'll attract scorn here but I like a continuous layer of external insulation. It's keeps the studs and OSB sheathing warm and away from the elements. 

 

Provided you have adequate management of air movement through the structure (airtightness) and very good control over internal humidity levels the wall will quite happily dry to the inside provided you don't add any more low  permability layers like internal vapour barriors. 

 

Unfortunately airtightness is rarely done well and internal humidity control relies on the long term occupants diligence so there have been numerous building failures, hence the caution urged above. 

 

In your case the U value of the extension is lightly to be neither here nor there in the grand scheme of things. Aim for Bregs with something robust, easily buildable and cheap. 

 

What really will make it feel more comfortable is good air sealing. Make sure this is done well. I would avoid PIR internally as half of it will end up in the skip as offcuts and you'll be searching for studs to hit while screwing through it. Also it's virtually completely vapour closed so any accidental moisture in the wall will struggle to dry either in or out. Far easier to bang up a few 2*2s with a nail gun as a service cavity and mineral wool is fast and cheap to install. 

 

I really like blown cellulose for TF but it may not be worth getting the man for a small area of an extension. 

 

Importantly everything above is off the shelf at every builders merchant in the country. 

 

Robust, buildable. Cheap. 

 

 

 

 

 

It's not that hard to hit the studs when going through 50mm of PIR. I had to go through 100mm to hit my rafters and that was a bit of a pain, but it was manageable. 

 

Wool is cheap and breathable but you'll end up with a thick wall build-up, and a lot of the money you've saved on insulation will go in to extra timber.

[Iceverge]

1 hour ago, Roger440 said:

Im assuming that you would put some sort of cladding on battening on the outside of that build up?

 

 

Yes. Didn't include it because I'm lazy. 

[sgt_woulds]

16 hours ago, ADLIan said:

Do not render direct onto (any) insulation onto timber (or steel) frame construction!! It's not accepted good practice and lots of failures reported. About this time last year I was dealing with an architect looking at remedial works on this form of construction - sole plates and base of studs all rotted on 2 houses built approx 15 years ago.

 

Does PUR really form a VCL if the board joints are not sealed/taped in any way? Moisture vapour is a gas so will readily move through the joints. Discuss.......

 

 

Sorry, that is outdated thinking based on old building standards.  It is perfectly acceptable to directly render onto a breathable insulation for a vapour open structure both here and in the whole of Europe. 

 

Building inspectors may expect to see a ventilated cavity based on previous experience, but point them to the manufacturers of such systems and they will be placated.  

 

Without knowing the construction details, your rotting sole plates indicates a host of other issues, (cold bridging, incorrect insulation type, poor internal VCL detailing).

 

STEICO (and other) woodfibre insulations use a hydrophobic coating on external sheathing boards.  In addition to the render it is more than adequate to keep external driven rain away from the structure.  Unlike unatural insulations, the woodfibre actively wickes moisture away when conditions allow.  

 

Most EWI systems use direct render onto the insulation.  Why should this be different for timber frame?

 

Any vapour open structure needs to be properly designed.  In Europe, TF houses are built inside out compared to UK; the OSB racking board is fitted inside the frame, not externaly.  In this location (with taped joints), the OSB acts as a moisture vapour variable control layer. 

 

If fitted externaly, OSB acts as a vapour check and requires an additional internal VCL, the correct sealing of which is vital to prevent issues.

 

The European TF vapour open building method removes any requirements for vapour control membranes, speeding the build and making it far less susceptable to issues caused by innatentive builders.  

 

 

 

16 hours ago, ADLIan said:

Does PUR really form a VCL if the board joints are not sealed/taped in any way? Moisture vapour is a gas so will readily move through the joints. Discuss.......

  

 

No, it needs taping and sealing as stated previously.   Untaped and in a poorly designed wall  it is the worst option, since moisture can easily get behind it, but not so easily get out again once it has condesed against the cooler structure.   Worst example of this is internal insulation using pre-insulated plaster boards where there is no way to properly seal the joints.

 

 

 

11 hours ago, Iceverge said:

Provided you have adequate management of air movement through the structure (airtightness) and very good control over internal humidity levels the wall will quite happily dry to the inside provided you don't add any more low  permability layers like internal vapour barriors. 

 

Unfortunately airtightness is rarely done well and internal humidity control relies on the long term occupants diligence so there have been numerous building failures, hence the caution urged above. 

 

 

 

Control layers of any kind are rarely done well by the Lesser Spotted Great British Builder.  I don't think we have had a breeding pair of Great Builders in the UK for a long time....

 

This is why a fully breathable structure using natural insulations that doesn't require perfectly sealed membranes just makes more sense. 

 

 

 

11 hours ago, Iceverge said:

In your case the U value of the extension is lightly to be neither here nor there in the grand scheme of things. 

 

 

 

I agree, decrement delay is far more important, espescially if it has a flat roof.   

Edited 6 hours ago by sgt_woulds
spelling

[SimonD]

1 hour ago, sgt_woulds said:

Sorry, that is outdated thinking based on old building standards.  It is perfectly acceptable to directly render onto a breathable insulation for a vapour open structure both here and in the whole of Europe. 

 

Building inspectors may expect to see a ventilated cavity based on previous experience, but point them to the manufacturers of such systems and they will be placated.  

 

Without knowing the construction details, your rotting sole plates indicates a host of other issues, (cold bridging, incorrect insulation type, poor internal VCL detailing).

 

STEICO (and other) woodfibre insulations use a hydrophobic coating on external sheathing boards.  In addition to the render it is more than adequate to keep external driven rain away from the structure.  Unlike unatural insulations, the woodfibre actively wickes moisture away when conditions allow.  

 

Most EWI systems use direct render onto the insulation.  Why should this be different for timber frame?

 

Any vapour open structure needs to be properly designed.  In Europe, TF houses are built inside out compared to UK; the OSB racking board is fitted inside the frame, not externaly.  In this location (with taped joints), the OSB acts as a moisture vapour variable control layer. 

 

If fitted externaly, OSB acts as a vapour check and requires an additional internal VCL, the correct sealing of which is vital to prevent issues.

 

The European TF vapour open building method removes any requirements for vapour control membranes, speeding the build and making it far less susceptable to issues caused by innatentive builders.  

 

I think you need to take a more nuanced perspective on this as the picture is not as wonderful as this makes out. There are plenty of reports of problems with moisture ingress and failure of these rendered board in certain climates across Europe. For example:

- All along the west coast of Norway, thin coat renders (directly in insulation) were suffering from premature failure as a direct result of wind-driven rain. As a consequence Norwegian building standards now have a map of areas where it is deemed unsuitable.

- In Finland it was found that thin coat renders directly on insulation suffered from freeze/thaw failures as in that climate it is common to have lots of rain followed by freezing temperatures at night. This failure is in part due to how the render experiences a thermal shock due to local thermal conditions with the insulation directly behind it. This is one of the reasons also why many lime renders will simply not be suitable for woodfibre in particular and you must choose a render with specific ingredients to deal with the behaviour of the background material.

 

I have woodfibre directly rendered, which sits on both masonry and timber frame, but being where we are, it's not exposed at all. If I were somewhere that freeze/thaw cycles were prevalent, like Scotland, and there was particular risk of wind driven rain, I'd be using a cavity and render board, no question.

 

And also, although the woodfibre is often treated with a parafin like substance to provide better moisture resilience, it is not only vapour open, but it is hygroscopic and therefore any designs must take that into careful account.    

[Gone West]

2 hours ago, SimonD said:

- All along the west coast of Norway, thin coat renders (directly in insulation) were suffering from premature failure as a direct result of wind-driven rain. As a consequence Norwegian building standards now have a map of areas where it is deemed unsuitable.

- In Finland it was found that thin coat renders directly on insulation suffered from freeze/thaw failures as in that climate it is common to have lots of rain followed by freezing temperatures at night. This failure is in part due to how the render experiences a thermal shock due to local thermal conditions with the insulation directly behind it. This is one of the reasons also why many lime renders will simply not be suitable for woodfibre in particular and you must choose a render with specific ingredients to deal with the behaviour of the background material.

IIRC @ProDave had a problem with render on wood fibre EWI, but I think it was a while ago, so I may have got it wrong.

[sgt_woulds]

True enough, and I should have made it clear that I was talking UK generally.

 

The Norway / Finland issues a combination of extreme differences between internal/external conditions for prolonged periods which added internal moisture to external external wet conditions with the added fun of sustained periods below freezing.  I'll see if I can find the report but I think they also blamed higher average internal moisture due to saunas and gas cooking, but I might be getting it confused with a similar case on some islands off the coast of Germany.  Surprisingly, the reports I've read include poor instalation which I had thought was a UK specific issue.

 

In the UK, except on the coast at the far North of Scotland these specific issues are unlikely to occur.  

 

It should be noted that, (unless it is open slatted) where an external ventilated rainscreen is used, an external membrane is still not required for hydrophobic coated woodfibre external insulation.  Although membranes are normally specified anyway as BCO 'expect' to see membranes... 

[SimonD]

12 minutes ago, sgt_woulds said:

n the UK, except on the coast at the far North of Scotland these specific issues are unlikely to occur.  

 

You see quite a lot of freeze/thaw in the UK in winter - we even get it inland in England, especially with a well insulated house. There are a lot of places along the whole of the west coast of the UK that can suffer from wind driven rain without the opportunity to dry out. With the Finnish issue it was not the internal/external conditions but specifically to do the damp air/rain wets the thin coat render, freezes overnight and then thaws again in the morning. This required a revision of thin coat render standards and testing for the Finnish market in addition to the standard European tests. The thermal shock requires different polymers to be added to the render.

[sgt_woulds]

Yes, but it was the freeze thaw that was the specific issue and the fact that there was no other route for the moisture to escape - the high internal humidity restricted flow that way too.  

 

While we may have high saturation rates along the UK coast, similar sustained freeze thaw cycles at the same are unlikely.  

 

Tested systems e.g. Baumit will take these extremes into account.  As ever, informed specification is key.   Seek the advice of the woodfibre and render manufacturers and build specific to the local microclimate.

 

This is different to saying:

 

21 hours ago, ADLIan said:

Do not render direct onto (any) insulation onto timber (or steel) frame construction!! It's not accepted good practice

 

 

 

[SimonD]

4 minutes ago, sgt_woulds said:

the high internal humidity restricted flow that way too.  

 

This will not in any way affect the local moisture conditions at the insulation/render/outside. It takes weeks if not months for moisture to pass through to the inside and with vapour pressure difference between outside driving vapour to the outside it's going to move in that direction. 

[Iceverge]

1 hour ago, SimonD said:

 

This will not in any way affect the local moisture conditions at the insulation/render/outside. It takes weeks if not months for moisture to pass through to the inside and with vapour pressure difference between outside driving vapour to the outside it's going to move in that direction. 

 Sadly @Gone West it was @ProDave who had the issues. Baumit silicone render over wood fiber I think I remember. 

 

That's assuming it's vapour driven moisture that's the issue. It's largely isn't in reality but rather the gaping holes left by the building process in the wall. 

 

I would be shy of rendering onto insulation in anything but a dry and stable climate.  Ireland certainly isn't.  I've seen another render failure over EPS near me recently. Constructed about 5 years ago. 

 

Meanwhile I've seen plenty of westerly facing sand and cement plaster over 60 years old on our farm with zero paint or maintenance and it seems to be holding on fine. 

 

For Timber frame especially I would like a cavity. Mineral wool or wood fiber being my preference for insulation external to the studs as they won't trap moisture. 

 

In any case I think for @Jess Shannon  you'll make very little difference to the performance of an old building by getting caught in this rabbit hole.  

 

My suggestion of a 170mm wall, airtight layer and 50mm service cavity stands. 

 

 

 

 

 

 

 

[sgt_woulds]

1 hour ago, SimonD said:

 

This will not in any way affect the local moisture conditions at the insulation/render/outside. It takes weeks if not months for moisture to pass through to the inside and with vapour pressure difference between outside driving vapour to the outside it's going to move in that direction. 

 

 

I think the point is that this was measured over months and with higher sustained internal moiture levels.   But it is pointless discussing technical points without the reports to hand.  My old grey cells aren't good enough to remember the details!

 

I'll ask one of our technical guys if he has a copy of the report - I haven't found the same one with an interweb search yet.  The one I'm looking for is specific to natural (hygroscopic) insulations which work differently to oil and mineral based systems.  It will have to wait until after the Easter break.

 

Like you I have directly rendered woodfibre (mine is over a shitty SIPs build), but I live in a very exposed location on top of a hill in one of the windiest parts of the country, so I'm happy to practice what I preach...