Benpointer Posted February 21 Posted February 21 (edited) Getting structure quotes in for our 180m2 single-storey self-build. We've got vaulted ceilings which is complicating things a bit. The quotes are varying between £400 and £800 per m2 to get to u-values of 0.12 walls, 0.1 roof. One quote I quite like is Turner Timber's Super Advanced, which seems like a good price, and uses 254mm Easi-joists with 11mm OSB over for the roof, so no purlins intruding into the ceilings and potentially lots of room to run MVHR ducting and other services through the Easi-joists. Question is: should we insulate Between and under the joist, which loses the service space benefit, or Go for a warm roof with all the insulation on top, or Between and over, say 100mm PUR between and 150mm over (is that still a warm roof)? The roof will be a standing seam zinc so if we go option 1, we'll need membrane, 50mm battens, 18mm OSB or ply on top. VMZinc say you don't need that for a warm roof so is that a saving? Of course, I'll be asking our architect, who I rate, but fear I could be paying him £££ per hour to research this out, so if I can steer him in the right direction... 😉 Any views? Edited February 21 by Benpointer
Benpointer Posted February 21 Author Posted February 21 Oh, some pictures: Floorplan: Roof truss plan: Cross-sections:
JohnMo Posted February 22 Posted February 22 (edited) Over is the easiest to do, but if not careful can make roof look a bit clunky. But you still need to insulate at the outside wall to bridge between wall and roof insulation. Between - Any solid insulation such PIR between is a bit of a non starter. How do you go between the metal webs. Blown insulation, tried to cellulose but they wouldn't do it for some reason. My roof is similar, I did counter battens spaced away from rafter 50mm to drop ceiling down 100mm, to give 356mm cavity and spray foamed. Then a reflective vapour barrier and 50mm service cavity then plasterboard. Drape the vapour membrane onto wall and use airtight mastic/glue to stick to wall or tape if a membrane. Edited February 22 by JohnMo 1
Alan Ambrose Posted February 22 Posted February 22 (edited) That style of roof is just calling out for blown in cellulose / eps beads / wood fibre no? What is the wall structure out of interest? Edited February 22 by Alan Ambrose
Benpointer Posted February 22 Author Posted February 22 3 hours ago, Alan Ambrose said: What is the wall structure out of interest? Wall structure is as per this diagram but we're cladding in cedar / cement boards & render rather than having a masonry skin. Either 50mm additional PIR for u = 0.12 or 100mm for u = 0.09.
Benpointer Posted February 22 Author Posted February 22 (edited) 4 hours ago, JohnMo said: Between - Any solid insulation such PIR between is a bit of a non starter. How do you go between the metal webs. So, I am thinking there will be a calculated u-value from, say, 100mm PIR between the joists with nothing within the joists (i.e. between the webs), then say 150mm PIR over the joists. Sure it will be less than full-fill 100+150 PIR but better than just the 150mm over the top. Presumably could be calculated for the whole roof by calculating the % area with PIR versus % of unisulated joist. For 97mm joists at 600mm centres that's 83% PIR at 0.22u and 17% 47mm joist at 2.55u = 0.6u... Then add the 11mm OSB and 150mm PIR on the top to get an overall u of 0.11. Still, it might be easier/cheaper to simply put 200mm PIR on top and forget insulating between the joists. I don't want to full-fill the easi-joists because I'd like to thread services including MVHR through them. I am mindful of making the roof look very clunky though. Edited February 22 by Benpointer
Benpointer Posted Thursday at 20:22 Author Posted Thursday at 20:22 On 22/02/2025 at 09:36, Alan Ambrose said: That style of roof is just calling out for blown in cellulose / eps beads / wood fibre no? What is the wall structure out of interest? We're now investigating blown cellulose (254mm), with a layer of PIR on the warm side to get us to u = 0.1. However, the Warmcell distributor (PYC) are recommending that we'll need Smartply Propassive on the inside and Medite Vent on the outside, which would add £20 per m2 to the roof cost and that's before we factor in the cost of the blown cellulose itself. Given we will also have to have a 50mm ventilation gap then 18mm ext ply above the roof to take the zinc we would end up with 3 timber board layers in the roof: Smartply Propassive, Medite Vent and 18mm ext. ply. That seems like overkill to me - not sure why we can't go with a suitable VCL below and 9mm OSB3 above the cellulose. Ubakus certainly seems to think it's ok: The alternative might have to me mineral wool between the rafters rather than cellulose, which is a shame as I like the idea of blown cellulose. (-:
JohnMo Posted Thursday at 21:25 Posted Thursday at 21:25 T 50 minutes ago, Benpointer said: We're now investigating blown cellulose (254mm), with a layer of PIR on the warm side to get us to u = 0.1. However, the Warmcell distributor (PYC) are recommending that we'll need Smartply Propassive on the inside and Medite Vent on the outside, which would add £20 per m2 to the roof cost and that's before we factor in the cost of the blown cellulose itself. Given we will also have to have a 50mm ventilation gap then 18mm ext ply above the roof to take the zinc we would end up with 3 timber board layers in the roof: Smartply Propassive, Medite Vent and 18mm ext. ply. That seems like overkill to me - not sure why we can't go with a suitable VCL below and 9mm OSB3 above the cellulose. Ubakus certainly seems to think it's ok: The alternative might have to me mineral wool between the rafters rather than cellulose, which is a shame as I like the idea of blown cellulose. (-: This sounds like the conversation I had about cellulose, they wouldn't even quote until I had all the drawings changed to suit what they expected. So zero idea of costs without going back to architect to amend drawings - so moved on.
Benpointer Posted Thursday at 21:38 Author Posted Thursday at 21:38 (edited) 14 minutes ago, JohnMo said: T This sounds like the conversation I had about cellulose, they wouldn't even quote until I had all the drawings changed to suit what they expected. So zero idea of costs without going back to architect to amend drawings - so moved on. Tbf they seem to want to insist that the inside layer SD value is >= 5x the outer layer SD value. Our outer layer is likely to be 11mm OSB/3 with an SD of 2.5. So we just need a VCL with an SD of > 12.5 or more. DuPont AirGuard Reflective has an SD of 2000 so should more than be sufficient. I have put that build-up to PYC to see whether they are ok. The problem is on Ubakus that build up gives an insufficient drying reserve.🤷♂️ Edited Thursday at 21:40 by Benpointer
Benpointer Posted Friday at 08:17 Author Posted Friday at 08:17 Also, just taping up the TP10 should give me a sufficient SD on the inside surely?
Recommended Posts
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 accountSign in
Already have an account? Sign in here.
Sign In Now