IWI and Condensation (PIR and Studwork)

[noseall]

ah OK, so you have stud work fixed to the wall, PIR friction fitted between them then battens fitted on top of the PIR for a service voice, and thin PIR around cabling.

RE polythene, do you install this on the cold side of the insulation i.e. on the wall or the warm side (under the battens).

No. The battens go on the outside of the first 50mm layer and are fixed through. It just ensures continuity of insulation, tight against the wall.

 

[mrrusty]

but we want the flexibility to be able to fill is retrospectively

I would seriously think about this. When we IWI our small cavity wall, I discussed with our BCO about his view on interstitial condensation. We both came to the conclusion that when it occurred the ventilation in the cavity would evaporate it. Our cavity is very well ventilated. And so it has proved - we have had no issues.

Early cavity walls were designed to be ventilated, and the original purpose was to keep the inner skin, and hence the ends of joists dry and rot free. The cavity was not there initially for insulation.

 

[mrrusty]

The beauty with separates is the joints can be staggered. There have been instances where the flex or memory of the bound boards has cause fine cracking, even with scrim. We now use the 100mm orange scrim on those boards and plenty of mechanical fixings (post dab).

We haven't had any issues, but I take the point. I would certainly take your suggestions on board (!) next time I do something similar. always interesting to hear thoughts from the more experienced people...

 

[timwagg]

thanks for everyones input, it's been most helpful. we've decided to leave the cavity unfilled and go for a PIR + PB approach, taping the seams and the join to create. a air tightness barrier alongside foam for any major unduliations in the wall behind (there isn't much as it's a block wall) and sealing the join at the floor boards with an air tight and flexible silicone or such like.

for the record, we aren't adding a vapour barrier in the roof as our research suggests any warm air that escapes will slowly cool through the 300mm of insultation that is going up there plus the air flow is abudant up there (will leave at least 50mm between rafters after we have push the PIR in between

one more question i do have...there is one remaining wall, used to be the external facing wall of the house (which dates back to 1800 and something). It's about 14 inches thick but is now essentially internal. It has been extended upwards with block work so the roof line can be raised.

normally I wouldn't consider insulation interally but the wall gets very cold in winter and it will form a wall in our new bathroom. It was previously a bathroom and was dot and dabbed with plasterboard, the wall behind is very dry but it did get very cold, far from ideal in a bathroom.

i understand that PIR is a no no when it comes solid walls on external faces but as this is technically internal can I proceed with a similar fixing and material. I'll prob go a bit thinner as it's internal but I do feel it needs something. for context this wall will also have a shower on it (but services will run elsewhere).

thanks

 

[mrrusty]

but the wall gets very cold in winter

If it's internal more likely it just takes a very long time to heat up. Stonework will have a huge heat capacity and although the room air will warm quickly, the wall will stay cold until enough heat energy is available for it to warm, then it will stay warm long after the room has cooled down - just like an old fashioned storage heater...

 

[robinbanks]

sealing the join at the floor boards

Floor boards are not renown for being airtight (it doesn't seem the type of house to have joint glued eggerdeck protect for example), so I'm not sure why you'd seal PIR boards to them. If you're going at it hard core you'd lift the floorboards, insulate between the joists, put a vapour barrier down and refit the floorboards over it, and seal the wall boards to the vapour barrier. Dedicated barrier systems with tapes and adhesives exist for this

we aren't adding a vapour barrier in the roof

It seems daft in my book, especially if the insulation you use there is vapour permeable, as the dew point will be somewhere inside the insulation, and if your warm moist air is transiting through it, that's the point at which it will condense

the wall gets very cold in winteR

Why? Is the other side of it unheated? Then either heat and insulate that space or insulate the bathroom against it (if eg it's a conservatory and basically follows the outside world temp with some small delay)

The simple notion here is you have to decide where the heated envelope of your house lies, and put insulation between that line and the world. The insulation is what allows a steep temperature gradient to establish eg 20 degrees inside, 0 degrees outside, the temp drops across the insulation (if you dug out half of it with a spoon and measured the temp the middle would be 10 degrees) and this is also where you get your condensation risk point- if that dew point is in the middle of a vapour impermeable block of insulation, condensation doesn't form there because moist air can't reach there

i understand that PIR is a no no when it comes solid walls

Why do you say this? Insulating a house externally is a perfectly valid approach and the walls warm up, bringing temperature stability (thermal buffering)
What you don't typically do is mix internal and external because then you just create a path for heat to flow out of the building- thermal bridge

thanks for everyones input, it's been most helpful.

So you're now clear that you're going to stick standard PIR to the wall with foam, tape the joints so the whole wall is PIR, then use through fixings long enough to go right the way through the PIR, to hold 25mm battens against the PIR (which also helps hold it against the wall), then you'll run your wires and pipes (notch the battens - I notch mine at set heights on a mitre saw before I attach to wall), then you'll insulate the rest of the service void by filling it with 25mm PIR, taking care not to closely insulate any wire runs, then you'll fit plasterboard over, taki care bit to drive any screws through any wires, and you'll use plasterboard back boxes for electrical points that are of a suitable depth (37mm if your pb is 12.5mm) so as to not need to dig a hole in the insulation. And you will run all your wires as according to safe zone rules?

Good stuff!

 

[timwagg]

Floor boards are not renown for being airtight (it doesn't seem the type of house to have joint glued eggerdeck protect for example), so I'm not sure why you'd seal PIR boards to them. If you're going at it hard core you'd lift the floorboards, insulate between the joists, put a vapour barrier down and refit the floorboards over it, and seal the wall boards to the vapour barrier. Dedicated barrier systems with tapes and adhesives exist for this

Understood, this is where I have been struggling to understand the importance of air tightness. This part of the house I am working on was built in 1995, and it's been done to a good standard but as you say, new glued eggerdeck - just t and g FBs. I've got to pull some of the FBS up anyway to run services so insulating down into the void below could be done but I thought it may just be easier to fill it with lof insualtion (currently planning Knauff Loft Roll 44. If I did cut the insulation to sit down into the void, I assume I'd tape using the same foil tape used to join the PIR together, but onto the joists? I read elsewhere that taping into joists isn't generally a good idea when working a roof space as it can stop the wood breathing, is there a similar issue here?

It seems daft in my book, especially if the insulation you use there is vapour permeable, as the dew point will be somewhere inside the insulation, and if your warm moist air is transiting through it, that's the point at which it will condense

I'll be honest to not add one is due to a few reasons 1. cost - using something like intellio would cost a fortune 2. if I used a polyethelene sheet, I'd need to put it on top of the joists and then insulate on top as I plan to keep my exisiting ceiling in place and board over, meaning faff bringing the vapour barrier down between each joist, sealing etc. will use knauff loftroll 44, 100mm between joists then 200mm onto at 90degrees.

Why? Is the other side of it unheated? Then either heat and insulate that space or insulate the bathroom against it (if eg it's a conservatory and basically follows the outside world temp with some small delay)

The simple notion here is you have to decide where the heated envelope of your house lies, and put insulation between that line and the world. The insulation is what allows a steep temperature gradient to establish eg 20 degrees inside, 0 degrees outside, the temp drops across the insulation (if you dug out half of it with a spoon and measured the temp the middle would be 10 degrees) and this is also where you get your condensation risk point- if that dew point is in the middle of a vapour impermeable block of insulation, condensation doesn't form there because moist air can't reach there

it's the original solid wall of the house which dates back to 18 something, but it has been built on top of with breezeblocks (I believe still solid) to raise ridge. the other side is heated but I think due to the sheer width of the wall (around 12 inches) it takes a long time to heat up, and other issues like drafts and works to be carried out mean it is hard to get the house to heat up and stay warm (hence why I am so paranoid about doing a good job here). my idea is is that by adding a thin layer of insualtion it will help keep the thermal comfort inside the bathroom especially as the wall will have the shower which will be tiled. When i was taking the PB (dot and dabbed) off earlier today the wall is bone dry, but was uninsulated.

Why do you say this? Insulating a house externally is a perfectly valid approach and the walls warm up, bringing temperature stability (thermal buffering)
What you don't typically do is mix internal and external because then you just create a path for heat to flow out of the building- thermal bridge

my understanding is that with IWI if it's not thick enough, if warm air gets through the insulation it can condense onto the cold wall. essentially from that standpoint it's external as it is unlikely to heat up to the amibient air temp. this can also happen the other way, I'm less worried about this on the other walls as they are cavity.

So you're now clear that you're going to stick standard PIR to the wall with foam, tape the joints so the whole wall is PIR, then use through fixings long enough to go right the way through the PIR, to hold 25mm battens against the PIR (which also helps hold it against the wall), then you'll run your wires and pipes (notch the battens - I notch mine at set heights on a mitre saw before I attach to wall), then you'll insulate the rest of the service void by filling it with 25mm PIR, taking care not to closely insulate any wire runs, then you'll fit plasterboard over, taki care bit to drive any screws through any wires, and you'll use plasterboard back boxes for electrical points that are of a suitable depth (37mm if your pb is 12.5mm) so as to not need to dig a hole in the insulation. And you will run all your wires as according to safe zone rules?

Good stuff

yep that's the long and short of it. I will speak to my spark RE: distances for cable runs. as the room is semi 'in roof' the plan is to fit 100 - 150mm insulation between rafters, then 50 mil ontop the rafter and brining that down to the 50mm fixed to the wall then taping them up. roof space as outlined above. one of the major reasons for wool rather than PIR in ceiling is for eay access to services and cost.