Why internal wall insulation (IWI) matters
To radically address carbon emissions in the UK we are going to have to undertake an extensive retrofit of existing buildings. External wall insulation is not a viable option for many buildings in the UK and so we will need to get to grips with how we use internal wall insulation (IWI) effectively and safely. There is currently much ongoing research and debate about the dangers of interstitial condensation when using IWI, with concerns about moisture and mould building up behind and within the installed insulation. The forthcoming AECB CarbonLite Retrofit programme is seeking to address these issues among other and share best practice from projects around the UK.
In many ways, as unsexy and unglamorous as it may be, thinking about the best ways to install internal wall insulation is at the current cutting edge of sustainable building in the UK. It is in this context that we are attempting to approach the IWI requirements at our Cumberworth radical retrofit project.
As we said in the last blog, the risk factor of interstitial condensation is different for different construction types. For this reason, our insulation strategies at the Cumberworth project are tailored to the different construction types.
Barn (solid stone, rubble filled)
For the barn internal walls, we are trialling for the first time in the UK, a new innovative insulation material called TecTem from Knauf AquaPanel in Germany. It is designed for internal insulation applications, is vapour open and deals robustly with the common moisture issues of internal insulation systems. The product is not, at present, available on the UK market.
A positive aspect of TecTem is that you can, in theory, apply up to 200mm thickness to the internal face of walls because it ‘breathes’ in the way that the original construction did. Its thermal conductivity is 0.045 W/mK,. We do not know yet the final costings, but it does go a long way to reducing risk factors of interstitial condensation. It is particularly useful in situations where masonry walls are most at risk of moisture build up, so with the help of AECB Carbonlite we have identified these areas. We have parged with a weak sand and cement coat the internal face of the external wall to help minimise wind driven moisture and to give a flat surface for adhering the insulation. We have used water inhibitor in the parging for the bottom 900mm to inhibit rising damp. The East wall is particularly damp for the bottom 300mm and so we have decided to use 600mm high Foamglas slabs as an impervious tanking and insulating solution.
We are very excited to be trialling a product that appears to offer a significant advance for this application, with the only main disadvantages potentially being cost and difficulty of applying it. It is a very delicate material, soft and quite crumbly. The product is already attracting a lot of interest from the AECB CarbonLite researchers, who have installed moisture monitors throughout the project, to monitoring moisture levels in the Cumberworth retrofit walls.
1990s extension (cavity wall construction)
The risks of moisture build up and interstitial condensation are not as great in the 1990s extension and so we have developed a different IWI strategy for that area. There had been significant water ingress during the re-roofing operations resulting in high moisture content in the plaster and masonry. Normally it would be expected that the moisture in the wall could dry easily through the block to the cavity, however the inner leaf was constructed with a foam insulated block. This type of block which was popular in the 1990s in Yorkshire was supplied with either expanded polystyrene or polyurethane bonded to one face which acted as cavity insulation. The presence of the plastic foam insulation will have affected the drying potential so modelling was undertaken using WUFI to check that the wet plaster and masonry could dry in an acceptable period of time. This moisture levels in this construction are also being monitored to compare to the prediction in the modelling.
We are leaving the existing gypsum plaster in place and parging (applying a weak sand and cement mixture) in areas where there are gaps in the plaster (eg between intermediate floors) and then using proprietary insulated studs mechanically fixed, with low thermal conductivity mineral wool insulation between, followed by an Intello Plus vapour control barrier, service void and plasterboard.
On site
Reducing thermal bridging in structural steel
Our structural engineer Stuart McCormick from SGM has been helping us rationalise the space in the top floor which was really badly laid out. Most of the first month on site has been taken up with altering the roof trusses to accept new doorways through. We’ve had to put big steels at the intermediate floor level to support the truss allowing us to chop them out so we could put doorways in.
Unfortunately however those steels create bad thermal bridges to the external wall. To reduce this we’ve used Foamglas surrounding the ends of the steel. This has a high compressive strength and is consequently often used in foundations to minimise thermal bridging.
Bill Butcher, Director, Green Building Store www.greenbuildingstore.co.uk
As a Building Surveyor, I will be interested to see how this performs. A normal solid brick or stone wall plastered internally can be subject to surface condensation. The reason I assume it is not a big problem is mainly that warm air within the building “soaks up” the moisture. Breathability helps too as the problem is much more acute where an impervious surface finish has been applied.
Having a permeable insulation layer internally would dramatically reduce the temperature of the structural wall. So unless the warm moist air in the room is prevented from reaching it (by Intello membrane, for example), I would expect condensation to occur significantly on the inner face of the wall.
I know there is theory and practice. Theory tells us that there is such a thing as interstitial condensation and a theoretical risk of condensation occurring in the middle of a layer of insulation. I have never myself seen evidence of it in practice. All my experience suggests that condensation will occur at the nearest cold enough surface, which in this case would be the internal face of the masonry wall. Let’s see what happens here.
Simon, I don’t disagree with anything you say. The two strategies are designed to deal with the risks of the two different wall constructions. The solid/rubble filled wall has the added complication of driven rain and rising damp. There should be less risk at Cumberworth because of the planned for high level of airtightness detailing, and a very efficient MVHR system giving lower relative humidity. Carbonlite have now installed over 50 sensors to monitor most details, and I excitedly (sad!) await the results.
Bill, the detail shown with the foamglass doesn’t seem to address the issue. The entire end of the steel beam needs to be isolated from the cold masonry wall by an insulated pocket built into the masonry before the steel beam is inserted.
Steel is so conductive that any portion of it which is in contact with cold damp masonry will stabilise at the same temperature as the wall. This means that the steel will constantly attract condensation as it passes through the IWI and into the warm internal air. You need a 3D thermal model to establish the thickness of insulation required to ensure that no part of the steel has a temperature less than 15degC in order to avoid the condensation issue. No amount of ventilation will eliminate this issue, even if it was somehow possible to introduce forced ventilation into the floor void.
Simon,
The photograph of the steel joist bearing doesn’t do the detail justice. The bearing of the steel is completely surrounded by FoamGlass for compressive strength whilst there is 100 mm of polyurethane positioned between the end of the joist and outer stonework.
Bill
One further question: have you undertaken a Wufi analysis of the new “breathable” insulation product with a moisture load introduced just behind the external masonry surface? A slight crack in the external pointing may be enough to admit rain. If you run the same model without the moisture load you can assess the sensitivity of the construction to rain soaking. This will also vary according to the orientation of the wall, so make sure to test separately for each orientation. In a further iteration you can apply a hydrophobic, but vapour open, surface coating and test the effect. It may not be necessary to coat all walls, if the orientation suggests a particular orientation is significantly worse than the others.
The make-up of the masonry wall will vary from place to place so it would be appropriate to model several iterations to test for pockets of adverse effects. Random rubble construction has through-stones which will have a lower temperature and less liquid moisture holding capacity than isolated external and internal stone leaves separated by large amounts – possibly 30% – of lime mortar and lose fill. Each of these will require a separate Wufi analysis.
If the wall is not drying each year, then the design has to be regarded as likely to fail. The only question is how long you have got before it does.
[…] bridge both into the wall and into the ground. What we’ve done is use 30mm thickness of the TecTem IWI at a width of 400 mm on the return, as an anti-condensation measure so it keeps the first part of […]
Simon,
Thank you for your further comments, I don’t disagree with anything you say. We have carried out extensive WUFI calculations on all the wall types and elevations. We have also carried out an extensive moisture survey pre design. We also have the luxury of over 50 moisture sensors throughout the different wall structures and joist ends allowing us the security of knowing whether ‘brick cream’ is necessary or not. With the barn wall being 450 mm thick with rubble fill plus drawing on our 40 odd year experience working with this local stone and the Pennine rain, we feel there is a low risk, but time will tell. Our experience is that even with the better ‘brick creams’ they will start failing after 10 to 15 years. Can we guarantee or rely on subsequent home owners to recoat after 10 years? A 225 mm thick solid brick external wall facing west of course would probably be a different story………………
Bill
[…] has already discussed the decisions surrounding the internal wall insulation (IWI) materials in his blog. I will cover it in a later episode; suffice it to say that we are […]
Hi Bill,
Can you sit a shs base plate on a foamglas
pad to use as a thermal break. Would this be strong enough to take a 4kn load?
Regards
Matt
Matt,
This is more of a question for your structural engineer. I know that with a barn conversion project we are currently looking at with a similar detail, we are having to use http://www.schoeck.co.uk/en-gb/home propriety thermal breaks……….unfortunately more expensive………..but it’s difficult to argue with a structural engineer when not a structural engineer!
Bill
Hi Bill
Can the foamglass, or similar, be used to deal with floor joist timber ends going into a cavity, when insulating internally?
In my case, am planning on using diathonite in my build, its a cavity wall house which had 70mm already added into cavity. Joist ends stop at end of internal skin and therefore touch the eps. What I need to understand is can I leave the joists as they are, or do I need cut them and hang them, or some half way house of foamglass?
I also have a single steel beam (unusual I think for 20s construction, upon which a bedroom wall sits). What would you do here?
Thanks for any thoughts!
Paul,
My opinion is that your timber joist ends are at a far lower risk of getting wet compared with a solid masonry wall. A cavity should prevent rain driven moisture and would indicate a functioning dpc preventing rising damp. With 70mm of insulation in the cavity you will not be getting interstitial condensation either.
The weakness of leaving the joists in the wall is the potential lack of airtightness. I would suggest either parging or using the Diathonite through the floor void see above. We would then use Contega SL airtightness to tape from the joists to the wall.
There might be a case of inserting FoamGlas around the end of the steel beam as above but having 70mm of insulation in the cavity keeps the steel ‘in the warm’ anyway.
Hope this helps,
Bill
Thank you so much Bill, your thoughts are much appreciated 🙂
Hi Bill, what sort of product do you use for the waterproof parge below DPC?
Thanks
Barry