Attic Roof Design, makeup to achieve 0.1

[SBMS]

We're in the process of putting together a design spec and discussions with the architect for our roof construction method, and insulation.

 

We have a 2 storey house, with rooms in the roof, and one part of the roof is vaulted from the first floor to the roof (so no bottom chord on that part).  We have a bit of balcony overhang as well.

 

We've decided to go traditional brick and block with a 200mm cavity with EPS beads.

 

There are two primary construction methods we're looking at for the roof:

 

1. SIPS roof
2. Engineered attic trusses

 

Both options probably require a steel ridge beam, although SIPS might get away with a glulam ridge (and glulam purlins - which are a nice feature exposed in the rooms).

 

SIPS roof

A local SIPS company make a 195mm SIPS panel which achieves 0.13 u-value, so this plus a 37.5mm insulated board, or even higher would get us the U-value needed. No issues with having to be accurate with insulation. Makeup is relatively slim @ 200mm as well.

The primary negative that I have seen for a SIPS roof is complaints of increased noise, particularly when raining.  Is this others' experience, would a 195mm panel reduce this, and is there an easy option to mitigate if it is?

 

Engineered trusses

If we go the engineered truss route then we have some decisions to make regarding how we insulate it. I've put the commentary (mainly negatives) underneath each insulation method:

 

1. A 150mm PIR board between the joists (either with or without an air gap dependent on membrane) with an insulated plasterboard underneath would get us to 0.1
   1. Concerns regarding fitting - we could use gapotap, but timber can shrink and there are fiddly bits around the wall plate junction to ensure thermal bridging is minimal
2. Blown in cellulose (with or without an insulated plasterboard underneath?)
   1. No real concerns with fitting, but would potentially need a much deeper joist? I don't know what a 200mm cellulose U Value gives, we could possibly go for deeper joists but then we're losing headroom potentially inside
3. Blown in EPS Beads (with or without an insulated plasterboard underneath?)
   1. Same as cellulose but probably the same questions/concerns - however easier as the walls are going to be EPS beads the same firm could do it
4. Rockwool/mineral wool batts (again, probably would need an insulated plasterboard underneath)
   1. Would be fitted by hand, so would need to be on top of the fitters
5. Spray foam is OUT - so will not use for reasons that have been done to death on this forum

 

Has anyone any advice or feedback on the above, their experience, and what they would look to do?

 

Thanks in advance

 

 

[JohnMo]

Other choice is posi rafter or I beam and glulam. You can also under draw with battens to get insulation depth required with further layers and just insulated with a blow in product.

 

Plus another, which is blown mineral wool.

 

Then an airtight vapour barrier fully taped and bonded to the wall to make airtight. If installing lights a service cavity.

 

17 minutes ago, SBMS said:

Spray foam is OUT - so will not use for reasons that have been done to death on this forum

For others reading - Spray foam is only inappropriate when used in the wrong place or wrong kind used (close cell instead of open cell). Ok when used correctly - Sarking boards (with gaps) breather membrane above. And vapour stop below, open cell foam only.

[Iceverge]

1 hour ago, SBMS said:

A 150mm PIR board between the joists (either with or without an air gap dependent on membrane) with an insulated plasterboard underneath would get us to 0.1

1. Concerns regarding fitting - we could use gapotap, but timber can shrink and there are fiddly bits around the wall plate junction to ensure thermal bridging is minimal

 

 

Don't put PIR between the rafters.  Waste, fire, noise difficulty fitting etc. 

 

1 hour ago, SBMS said:

Blown in EPS Beads (with or without an insulated plasterboard underneath?)

1. Same as cellulose but probably the same questions/concerns - however easier as the walls are going to be EPS beads the same firm could do it

 

Don't do this, noise, fire, decrement delay all average

 

1 hour ago, SBMS said:

Rockwool/mineral wool batts (again, probably would need an insulated plasterboard underneath)

1. Would be fitted by hand, so would need to be on top of the fitters

 

 Maybe, good for noise, decrement delay and fire but itchy to fit and doesn't do anything extra for airtightness.   

 

1 hour ago, SBMS said:

Blown in cellulose (with or without an insulated plasterboard underneath?)

1. No real concerns with fitting, but would potentially need a much deeper joist? I don't know what a 200mm cellulose U Value gives, we could possibly go for deeper joists but then we're losing headroom potentially inside

 

 

Do this. noise, airtightness, fire, decrement delay all good. 

 

 

Do you have any cross sectional sketches and I can have a look at a good buildup?

[SBMS]

16 minutes ago, Iceverge said:

 

Don't put PIR between the rafters.  Waste, fire, noise difficulty fitting etc. 

 

Don't do this, noise, fire, decrement delay all average

 

 Maybe, good for noise, decrement delay and fire but itchy to fit and doesn't do anything extra for airtightness.   

 

 

Do this. noise, airtightness, fire, decrement delay all good. 

 

 

Do you have any cross sectional sketches and I can have a look at a good buildup?

Thanks @Iceverge - any reason you'd discount SIPS?

 

Will Elevations drawings suffice?

 

 

[SBMS]

2 hours ago, JohnMo said:

Other choice is posi rafter or I beam and glulam. You can also under draw with battens to get insulation depth required with further layers and just insulated with a blow in product.

 

Plus another, which is blown mineral wool.

 

Then an airtight vapour barrier fully taped and bonded to the wall to make airtight. If installing lights a service cavity.

 

For others reading - Spray foam is only inappropriate when used in the wrong place or wrong kind used (close cell instead of open cell). Ok when used correctly - Sarking boards (with gaps) breather membrane above. And vapour stop below, open cell foam only.

Posi Rafters look interesting... They'd be useful for the vaulted section.  Do they typically rest on a ridge beam and purlins, without the need for a bottom chord?

[JohnMo]

Photo from inside and outside, no bottom cord. Come ready cut to slot into place.

[SBMS]

51 minutes ago, JohnMo said:

Photo from inside and outside, no bottom cord. Come ready cut to slot into place.

Thanks @JohnMo - really interesting.  A portion of our roof is vaulted and the other portion isn't - how would you incorporate a second floor in that instance? Would it preferable to hang it from the internal block, or sit it on the wall plate, and then sit the posi joists on top of the floor?

[SBMS]

Just for reference, this is the elevation with some measurements for the roof sections:

 

 

 

[Conor]

We used thermohouse EPS roof panels. 250mm thick EPS. Can remember the U value but wouldn't be hard to fix on another 50mm PIR on the inside before your service void.

[JohnMo]

If you look at the image (previous posted) there is a raised platform at the back of the house. The wall in front of it is a structural wall. There were wall plates added to that and the rear wall to form a loft (plant room). We used something like 10 or 12 X2 timber, to form the floor and glued down tongue and groove 22mm waterproof floor panels as that is the only upstairs bit in the house.

 

 

 

[SteamyTea]

5 hours ago, SBMS said:

don't know what a 200mm cellulose U Value gives

Just the cellulose will be ~0.175 W/m².K.

 

[SBMS]

4 minutes ago, SteamyTea said:

Just the cellulose will be ~0.175 W/m².K.

 

Thanks @SteamyTea

@Iceverge would you ever recommend this plus an insulated plasterboard? I know you seem to favour Rockwool if more insulation is needed?

 

Still intrigued that no one has advocated SIPS?

[SBMS]

5 hours ago, Conor said:

We used thermohouse EPS roof panels. 250mm thick EPS. Can remember the U value but wouldn't be hard to fix on another 50mm PIR on the inside before your service void.

Was yours an ICF build?

[Iceverge]

9 hours ago, SBMS said:

Thanks @Iceverge - any reason you'd discount SIPS?

 

Will Elevations drawings suffice?

 

 

 

Sips are inflexible in a design sense and lightweight insulation in the roof is poor for heat protection and noise. 

 

They can work but you're tied into a crane and a chosen supplier. 

 

I would cut that roof and use blown cellulose. 

 

 

 

 

 

 

 

 

 

 

[Conor]

7 hours ago, SBMS said:

Was yours an ICF build?

Yes.

[SBMS]

10 hours ago, Iceverge said:

 

Sips are inflexible in a design sense and lightweight insulation in the roof is poor for heat protection and noise. 

 

They can work but you're tied into a crane and a chosen supplier. 

 

I would cut that roof and use blown cellulose. 

 

 

 

 

 

 

 

 

 

 

Would trusses be out due to the intersecting roofs? Do you think posi rafters could work?

[Iceverge]

Trusses are perfectly fine apart from the trickeries of making them airtight. 

 

Posi rafters or I joists would work fine. Weigh up the extra cost of them Vs sawn lumber however. 

 

A layer of something like woodfiber outboard with a fluffy insulation might be cheaper than an engineered product. 

[BadgerBadger]

19 hours ago, SBMS said:

Would trusses be out due to the intersecting roofs? Do you think posi rafters could work?

 

Send your plans over to a couple of truss companies and see what they can do.  I found them very helpful.

 

We had a semi-vaulted bit of our build and the architect/SE were looking at using a large steel ridge beam.  Truss company had a look, made it work and advised how high we could raise the truss tie in the design before it started to hit their limits. Much simpler. (although I'm still faffing about with insulation and airtight tape on it!)

[SBMS]

5 hours ago, Iceverge said:

Trusses are perfectly fine apart from the trickeries of making them airtight. 

 

Posi rafters or I joists would work fine. Weigh up the extra cost of them Vs sawn lumber however. 

 

A layer of something like woodfiber outboard with a fluffy insulation might be cheaper than an engineered product. 

Thanks @Iceverge.  
 

Would posi rafters be able to span from the ridge to the wall plate without a purlin (or supporting upright like an attic truss). With rooms in the roof/ part vaulted ceiling that would provide a better internal aesthetic. 

[SBMS]

1 hour ago, BadgerBadger said:

 

Send your plans over to a couple of truss companies and see what they can do.  I found them very helpful.

 

We had a semi-vaulted bit of our build and the architect/SE were looking at using a large steel ridge beam.  Truss company had a look, made it work and advised how high we could raise the truss tie in the design before it started to hit their limits. Much simpler. (although I'm still faffing about with insulation and airtight tape on it!)

What is your airtight detailing at the wall plate? 

[JohnMo]

37 minutes ago, SBMS said:

Would posi rafters be able to span from the ridge to the wall plate without a purlin

Yes don't see it being an issue. Our longest are just over 7m between support. 

 

Airtightness details are super easy. We just covered the inside of the roof with airtight vcl and overlap on to the wall and then bonded to wall with airtight mastic. Then battened for plasterboard.

 

[Oz07]

13 minutes ago, JohnMo said:

Yes don't see it being an issue. Our longest are just over 7m between support. 

 

Airtightness details are super easy. We just covered the inside of the roof with airtight vcl and overlap on to the wall and then bonded to wall with airtight mastic. Then battened for plasterboard.

 

Did you counter batten below this for services or were they all coming up through the floor?

[JohnMo]

2 minutes ago, Oz07 said:

Did you counter batten below this for services or were they all coming up through the floor?

Full story 

Posi rafter, above was all sarking boarded, breather membrane and slates on 45 Deg roof and Sarnafil membrane on 12 Deg roof.  Inside was under drawn with counter battens to increase the depth from 256mm to 356mm. Full filled with open cell spray foam. VCL double sided taped to battens, stapled and all staples and joints aluminium taped. VCL draped down walls and bonded to wall. Counter battens (50mm service cavity for lights) then plasterboard.

 

When I battened the posi rafter, I did it to get a good thermal break. So used a small section of batten on a long section to form a 100mm 

 

Wall drape can be seen on wall.

 

Counter battens 

 

[BadgerBadger]

1 hour ago, SBMS said:

What is your airtight detailing at the wall plate? 

 

Just looking at this yesterday! Planning to parge coat top of blocks to get a smoother finish, then masonry airtightness tape (eg Solido SL) to seal between wall and ceiling insulation foil. Wet plaster to finish on walls.  No idea if it's a perfect solution or not, but seemed pretty good for the price of a couple of rolls of tape.

Edited June 23 by BadgerBadger

[JohnMo]

3 minutes ago, BadgerBadger said:

Planning to parge coat top of blocks to get a smoother finish, then masonry airtightness tape (eg Solido SL)

We used this on the house (parge coated walls)

https://passivehousesystems.co.uk/product/gerband-fortax-6400-airtight-sealant/

 

and this on the summer house

https://novia.co.uk/tapes/airseal310-sealant

 

Both seem to be exactly the same.