Jump to content

Combining EWI and Cavity wall construction on a New build


Iceverge

Recommended Posts

 

A friend came to me recently asking for advice on their new build . Its a nice design but, as they wish to push as far as possible towards passivhaus performance, there are some problems with thermal bridging. 

 

Specifically as shown below (hatched) where the architect drew external walls that would require internal support, steels and thus lots of thermal bridging. ( Some architects don't believe in physics....) 

 

 

image.thumb.png.b7399f8a02fc427f110655f4954a722f.png

The windows and long span ( >9m)  prevent a long Steel support holding up the external leaf independently from the internal wall. It wouldn't be very stable in any case. Internally insulating the steel has been suggested but it would still compromise the performance with high levels of insulation elsewhere. 

 

Would there be any issue with EWI only on this area and cavity walls elsewhere? 

 

I have suggested TF or  EWI for the whole house but their preference is really leaning toward cavity walls for cost, durability and familiarity reasons.

 

I assume they could be interfaced satisfactorily with a few appropriate plastering beads and applying the same acrylic render to both sides of the joint. Has anyone any experience of this? 

 

Maybe @nod or @Gus Potter are two that pop to mind. 

Link to comment
Share on other sites

It's a bit whacky for a domestic building but you can use what is called a Vierendeel Truss embeded in the inner leaf / as the structural frame . The base sits about the level of the ground floor ceiling... still in the insulation envelope, the top carries the roof loads. There is a bit of a cold bridge in terms of the inner skin but you can compensate for that by using insulated plaster board etc and treating the steels a bit like the inside of a window frame.

 

You can then use a shelf angle to support the outer skin. Now if you have a 9.0m span this could be economic.

 

But the shelf angle introduces a nasty twisting effect into the truss so you need to span the first floor joists perpendicular to the truss and use the joists and their connections to resist the twisting effect. It all takes a bit of thought and you need to get a grip on the how this type of structure behaves early on. If you plan it right it can be a very elegant way of doing it. These frame are usually made from square box section steel but for this span and the relatively light loads it needs to carry the section sizes can be quite modest with a bit of discrete stiffening.

image.png.981cd61138d63b93a254fdb71e275af9.png

The above is a basic one but in a fair wind you can adjust the vertical struts a bit to coincide with the internal walls and window openings.

 

The other main issue is lateral stability of the building. I can't make any more meaningful comment without seeing some floor plans etc.

 

 

 

 

Link to comment
Share on other sites

19 hours ago, Gus Potter said:

 

 Thanks for that Gus. Never heard of that truss before. Every day is a school day.

 

My friend is actually a SE so really well versed in making stuff stay up, just trying to help with some pointers I learned building our passivhouse.

 

In any case I modelled a block external leaf in THERM supported by an internal steel beam and provided you take care to insulate the rising block work externally it isn't actually too bad. 

 

I'll try to post if it I can get 5 mins at my laptop. 

 

 

Link to comment
Share on other sites

On 26/01/2024 at 16:38, Iceverge said:

My friend is actually a SE so really well versed in making stuff stay up

Spot on. The truss I show needs some adaptation to make it cost effective for the light loads and thin heavily insulated walls that you encounter in a domestic job but with a fair wind an a bit of lateral thinking / understanding of this type of structural truss it can be the perfect solution on occasion. It's not an unusual concept as we use this from time to time on tall buildings and you see it all the time on motorway foot bridges etc.

 

One of the big costs of these type of trusses lie in the welding cost. No fancy welds that need tested (keeps in the realm of the local fabricator), also don't skimp on the steel section wall thickness at the early design stage as later you could get tripped up when you come to the do the energy performance calculations.

 

Would be interested in hearing what you SE pal has to say and if said has a better idea.

Link to comment
Share on other sites

I'm not sure what the concern is - if the steels were dropped a bit then they'll be well within the thermal envelope. 

 

If you are concerned about the thermal bridge through the outer leaf then make the first few courses out of thermoblocks (the blue bit).

 

It depends what the internal detail is on how challenging the solution is. Bringing in a different wet trade for rendering a small section does not sound very economical. 

 

image.thumb.png.4d0d36d6d38406b4fb0c901edb868e7f.png

Link to comment
Share on other sites

The above is pretty much the conclusion I came to too. 

 

2-3 rises of thermal blocks and externally insulate them with 100mm of EPS or PIR inside the slanted roof and use 450mm of insulation on the flat part at the bottom of the triangle. 

 

 

Link to comment
Share on other sites

  • 3 weeks later...

image.thumb.png.7b6927b25faf3d4485213d125e3c62da.png

image.thumb.png.6b40a5067ad0bf24982e74b4e347296f.png

image.thumb.png.877ae2ac0d48f085fdd8cce556f48147.png

 

 

Here's a couple of screen grabs of that detail from THERM

 

300mm Cavity. 215mm internal block 100mm external.  Filled with EPS beads.

 

The initial 2 courses of blockwork on the external leaf are aerated concrete and externally insulated with 100mm of PIR. The insulation on the flat ceiling is 400mm of mineral wool. 

 

203UB  supporting both leafs. 

 

The higher the external insulation can be taken up the external leaf the better. However the thermal bridge vs a perfect imaginary steel free junction is about 0.03W/mK. 

 

So if you have this junction for 5m at dT20 then it'd be 5*20*0.03 = 3w difference in the heating load vs a perfect junction. Pretty acceptable in my view. 

 

 

 

 

 

 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...