SIP over cellulose?

[Alan Ambrose]

Hi,

 

Would anyone care to comment on this insulation stack-up, which I am considering for both walls and roof (walls with horizontal board for rain screen rather than pantiles)? Essentially SIP over cellulose using posi joists as Larsen truss. The idea is to use some PIR to reduce the overall wall thickness while maintaining a reasonable decrement delay / phase shift. This is all meant to be hanging outboard of an Oak frame which is also the structural support.

 

BTW I chose Isocell as the nearest thing to Warmcel / I think 'Lamination' in ubakus is probably ply / I don't know why ubakus looks like it has put in an extra layer of wood underneath the pantiles - but I don't think it makes any difference anyway as the cavity beneath is ventilated. I was also thinking of using the posi joists / Warmcel as service layer rather than have a separate one. I've included an extra ply layer inboard of the plasterboard (like our Scandinavian brothers) to allow easy shelf support etc.

 

It still looks there's a condensation problem - any ideas? Is the optimum decrement delay / phase shift 12 hours (for max attenuation) or is a longer number better?

 

TIA, Alan

[IanR]

It's breaking the rule of having increasingly vapour open materials towards the cold surface, hence the condensation at the PIR.

Unlike your section, SIP on the roof will have a lot more timber structure and therefore "cold bridging" within the SIP, and more condensation than shown,

Why not just go with 300mm or 350mm I-Joists, full of cellulose and a fibreboard external layer. Without any cold bridging, it would perform better in practice.

Edited November 20, 2023 by IanR

[Alan Ambrose]

Thanks Ian for your thoughts.

 

>>> Why not just go with 300mm I-Joists, full of cellulose.

 

A bit like this? I can get within 8% of the previous number with 373mm joists and only lose a cm:

 

 

I read the chapters on ventilation and moisture in The Passivhaus Handbook over the weekend, but I am obviously still not quite getting it...

 

BTW does this kind of join handle the SIP cold bridging problem?

 

 

[IanR]

10 minutes ago, Alan Ambrose said:

>>> Why not just go with 300mm I-Joists, full of cellulose.

 

A bit like this? I can get within 8% of the previous number with 373mm joists and only lose a cm:

 

Yes, more like that. Maybe still some work to do on I-Joist spacing. 400 centres may reduce the performance a little.

 

12 minutes ago, Alan Ambrose said:

BTW does this kind of join handle the SIP cold bridging problem?

 

 

No, I'm thinking of the extra timber content that gets included once they actually Engineer the panels for your roof.  Assuming the SIPs will be structural, there will be timber sections running the length of the SIPs to help take the load, and these create cold-bridges through the panel.  I did have a SIPs structure costed for my own build and my experience was that no one tells you about the performance effect until the last minute. 

 

Random picture off the internet to show what I mean. The SIPs roof engineered for my build had quite a bit more timber that shown in the image, but I do have a shallow pitch and 6m spans.