Attached is "standard" detail from Kingspan showing the principle of making the insulation layer in a conventional pitched roof continuous with that in a cavity wall. Essentially the top of at least the inner leaf needs to drop to a level at or below the bottom of the rafters, so that some form of insulation can cross over it.

However ... this seems to conflict with most standard details for the installation of restraint straps - for example those shown here:

https://nhbc-standards.co.uk/7-roofs/7-2-pitched-roofs/7-2-8-restraint/

which assume that the inner leaf rises to be level with the *top* of the rafters.

And I think that's to do with making sure the downward-turned arm of the restraint strap gets a good enough hold on the blockwork.

Is there a commonly accepted way of dealing with this conflict? for example using restraint straps with a longer downward arm?

Posted By: djhSorry but when I go to that link, I see a lot of stuff about lateral restraints but nothing about vertical ones? And the illustrations seem to show restraints under the rafters?

Yes, it is lateral restraints that I'm talking about.

The illustrations show the restraints fixed to the underside of rafters, but the inner leaf built up to the level of the tops of the rafters (not immediately obvious, but is made clear if you look at the 3d one in 3d, or look at the position of the "omitted for clarity" bit of blockwork relative to the strap).

I suspect the weight of the wall above the point where the strap is connected is what stops that connected block from simply toppling sideways. I'm surprised that as little as half a block is sufficient, TBH. If you could somehow reinforce the top of the wall to give it sideways strength maybe that would work. Longer straps as you say, perhaps on both sides? Dunno really; our structure is timber. Put a timber wall plate on top of the cut-down inner wall? And strap that down to the wall.

I am increasingly worried about the thermal bridging of gable wall inner skins from inside the insulation barrier to outside it.

I have, yes, stop the inside skin at ceiling level or take it down to that level.

On my new house no one said anything, single skin gable well tied to roof structure, no problems and no thermal bridging

Top of the gable retained by a ‘flying’ rafter, blocks wall tied to vertical framing (wind posts) in easy places and straps at 2m c/c screwed and plugged to blocks

I cut-in foamglass blocks into the inner leaf to deal with the thermal bridge at ceiling level (with EWI). But it's a huge faff. Still possibly easier than removing the whole inner leaf and adding windposts plus a way to hold up and restrain the purlins. My (1960s) roof doesn't have any lateral restraints except that the purlins are cemented into the wall. Not that much restraint there. It's not fallen off yet.

I do agree with Tony that the inner leaf bridging is quite significant with EWI (and generally ignored as it's hard to fix).

I think builders make the inner leaf the same height as the outer because it's just easier to cut it off there (and yes it helps support the tile battens). I'm not at all convinced that an extra ~120mm of block on top makes much difference to how well the restraint works.

Let’s have a little think about what we are trying to achieve. To stop the gable wall falling or getting blown in or out. I am sure that we have all seen cases of them falling out.

I like an exact 47mm gap between the first rafter and the blockwork. Then off cuts of 4x2 can be used as blocking pieces at the requisite centres to stop the gable falling in, no straps needed . Then on the outside if there is a barge board I like a backing rafter as it is very useful to nail battens to, if this either touches the outside skin or sits 47mm out then again blocking pieces can be used trapping the top of the whole gable.

So long as there are wall ties in the gable then it is now trapped and can’t fall over even without straps!

I never much liked noggins to fix straps to, preferring to nail on a chunk of wood and fix the strap to that or to any lateral braces, purlins etc. Sometimes we used to butt that against the blockwork and hook a strap over the block trapping it between the end of the wood and the bent bit of the strap, no need to fix to the block only fix the strap to the bearer. We think such a strap can be considerably shorter as the wood is connecting it to the rafters.

The same thing can be done at ceiling level - I like fail safe.

...