Posted By: Ed DaviesAh, so you were fibbing; the VCL is on the inside. Fine.

Posted By: borpinNo the celotex (foil faced PIR is outside the frame.

The frame's irrelevant. The key point is that what you're regarding as the VCL is on the inside of (enough of) the insulation.

I can't imagine how you're going to bond and tape the inside of the PIR properly with the frame in the way in places. Do you have a cunning plan? Is that what you mean by “Weak points are between the TF frames to ensure a good seal”?

By the way, my plan is much as Timber says, inside-to-out: rafters / (ply|OSB|T&G) decking / taped & sealed VCL / (PIR|PUR) / OSB sarking / (steel|oduline|whatever) / PV|ST

Ed/Timber,

I apologise if I misled you, I wasn't trying to fib; I was trying to illustrate the need to do without a 'conventional' VCL, as the inner (and outer) foil faces of the Celotex forms the VCL.
The outermost part of the TF is still the OSB sheathing, the outside of this is adjacent to 100m PIR, so this keeps the OSB and everything inside of this at near room temps/humidity, even in Dec and Jan according to the JPA Dewpoint analysis.

BC didn't seem to have an issue with it, and like I say the owners should be in soon....

Cheers..

Posted By: borpinThe frame is not irrelevant. The VCL is there to prevent moisture forming in an area that would be detrimental to the structure.

Indeed, you really don't want moisture and rot on the structure. But you don't want it anywhere else, either; the VCL is there to prevent significant quantities of water vapour getting to any cooler parts of the building.

Your buildup has OSB on the outside, mine doesn't.

Yes. It's a bit distracting in that I'm actually taking about my roof as I don't plan to have walls (A-frame). The outside OSB is the sarking board. Agree that for a wall in Scotland counter-battens would be fine but I think they get sniffy about those in roofs in windy places.

This is my point - really examine *why* you have the VCL - what is it actually doing?

I don't see how to seal the joins in PIR foil facing reliably when placing the sheets up against some other structure, be it a timber or steel frame or some sort of decking. It seems to me that laying some sheets, stapling it down and taping the joins so the staple holes don't leak is much simpler and more reliable.

Why do you need the inner OSB? Racking?

Yes, racking I imagine (to be determined). Also in my case it will be the finished wall/roof surface hence the probable use of T&G for at least part of the wall/roof just to look nice.

Why the external OSB? why not just counter battens? Where would the moist air come from and go to and where would it meet a cold surface? Your VCL is stopping moist air movement to where exactly - the inside of the PIR (as it can't go through it except in very small quantities even if you don't bond the PIR and tape it)

I'd have thought massive amounts of moisture (relatively speaking and over a long period of time) would migrate through the gaps between sheets.

So this air reaches the inside of the PIR, will the moisture condense? unlikely but remember, if it does it would condense on the inside of the VCL as that is the same temperature.

But the outside of the PIR is at much lower temperature. If the vapour gets all the way to the outside or even to half way out (e.g., with two layers of 130 mm PIR appropriately staggered) it'll condense.

My point is we seem to blindly follow the VCL dogma. In a first gen TF house (Studs, OSB and filled with rockwool) it was very necessary (in theory) as any moist air _would_ condense on the cold OSB. The construction of TF (and structure in general) has changed yet our thoughts on a VCL seem not to have.

You're right: it's the outer OSB layer's lack of permeability which is the problem. In many ways I'd rather do a pure “breathing” roof (as per Neil May) but cost and concern about fixing through such a thick layer of insulation put me off.

Still, I'm mostly interested to know how you can make the inside surface of the PIR really vapour tight when it's assembled against a frame.

Posted By: Viking HouseThese 2 studies may be of interest!
http://www.viking-house.ie/downloads/Tsong79.pdf
http://www.viking-house.ie/downloads/Spokane.pdf

Thanks - read the whole of the first and the abstract and conclusions of the second.

However, they're hardly discussing high levels of airtightness. In particular, the airtightness on the outside of the insulation is practically non-existent with near-air/vapour-open cellulose and mineral wool and poorly installed U-F with major shrinkage and cracking problems. They mention the use of felt as a moisture barrier but don't indicate how well sealed it typically was. I get the impression it wasn't that tight, though. This means that any water which does condense will have ample opportunity to dry out pretty rapidly.

This is supported by their observation of large numbers of water leaks of various types (through the cladding or from pipes) not causing much damage.

I think it would be courageous to extrapolate those findings to a wall or roof which has an airtight barrier on the outside, even if that barrier is supposedly vapour permeable. Also, I can't see how it helps with the case of PIR sheets which are pretty impermeable, I think, but have the potential for airpaths between the sheets if they're not joined properly.

Just read/re-read this thread from the beginning. I'm afraid we've had a bit of hijack (which I've aided and abetted) - maybe it would be worth going back to mattwprice's resurrection of this old thread two days ago to consider EWI retrofit on walls where at least the external leaf is brick.

Am I mis-reading the advice here? For our new house project I'm looking at around 200mm glassfibre insulation internal, including a 75mm internal wall / services space - 150mm timber frame fully filled and 50mm semi-rigid glassfibre external of the frame between timber cross battens to keep the frame warmer. There will be a vapour control membrane on the outside of that lot, mainly to give weather-proofing while we clad with T&G boarding and as the protection layer for any water getting past the boarding.

Basically it is a "breathing"wall construction as the house will be air-tight (using the OSB sheathing on the frame as the air-tight layer - either internally or externally depending on how we source the frame) and the ventilation system keeping the humidity of the interior to acceptable levels.

Is the advice in this thread moving away from the breathable construction? Thanks, Jonathan

Cerisy - Please make sure you get this correct.

You refer to a vapour membrane on the outside of the frame and insulation. This is a LOW moisture vapour resistance breather membrane. On the INSIDE of the wall will be a HIGH vapour resistance Vapour Control Layer (VCL).

PLEASE make sure you get this the correct way round.

For your wall construction, please IGNORE this thread!!

If you need further clarrification please do say. I would be happy to do you a drawing, to make sure there is no doubt!

In addition, I would recommend the use of a service void on the inside of the wall (inboard of the vcl) as this will allow you to achieve a much better air tightness performance.

Posted By: CerisyBasically it is a "breathing"wall construction...

...do you think I can fix the VCL to that...

Have we a terminology problem? Isn't a wall with a (working) VCL, by definition, not a “breathing” one?

David - that's for the response - the difficulty with fitting VCL on the inside of the 120mm semi-rigid insulation is how to fix it. The sequence is install the insulation and then the internal wall frames, so it would be impossible to fix the VCL on the back of the frames and seal the joints, etc. Hence the proposal to fit it to the OSB. Originally I had looked at omitting the VCL and allowing the moisture levels to balance out - a breathing wall concept as recommended by Pavatex - diffusion open as they call it! With the humidity levels in the house much reduced by the ventilation system the risks with condensation are also much reduced. Just questioning the decision now!!

Cerisy

As others have said, putting the VCL in the middle of the wall is a bit risky really, and may cause condensation.

What you are constructing, essentiall, is a twin stud wall construction.

How far along the process are you? Could you do something like this....

Cladding
Cavity
Breather membrane
OSB (sealed at joints to provide an outer air tightness layer
Load bearing timber studwork filled with insulation
Cavity filled with insulation
Second non load bearing studwork wall filled with insulation
High vapour resistance VCL membrane laped, taped and sealed to provide inner air tightness layer
25-50mm battens forming service void (also filled with insulation)
Plasterboard

That sort of gets you what you are after, but with a slightly more conventional set of layers and possitions of materials.

Ideally - you would construct the wall the other way round like this.

Cladding
Cavity
Breather membrane
OSB (sealed for outer air tightness)
Non load bearing timber frame (filled with blown in insulation)
cavity (filled with blown in insulation)
VCL
OSB
Load bearing timber frame (services between studs and insulation between)
Plasterboard

The second wall means that the thermal bridges at junctions are reduced, as well as keeping the structural frame on the inside of the insulation, and subject to internal air.

How far along with the project are you????

Worth considering air tightness junctions at ground floor, intermediate floor and roof junctions now. That may also dictate where the VCL sits within the construction.

For ease, the first of the two alternatives is better (easier to construct).

Blown in insulation in the second option. I wouldn't want to rely on just a membrane to hold it in. In the second option, the outer OSB could be substituted for a wood fibre board or bitumen softboard. It just needs something solid to hold it in place.