Airtightness in Retrofit – Parge coat under EWI

[Ruairi]

Hi All

In a number of retrofit projects we are doing at the moment, we are applying EWI over existing brick solid walls. I know on Andy Simmond’s PH retrofit and I think Alan Clarke’s as well, they applied a parge coat to the existing brick/stone to provide an airtight barrier then at roof level the airtightness layer was applied on top of the existing roof joists (roof stripped back). Then the new rafters and insulation were fitted on top. This all makes perfect sense as the airtight layer can be continuous and the new roof overhang covers the EWI.

Our projects are slightly different as we will only be stripping back a few course of tiles at eaves and we will be insulting the roof at ceiling level.  The air tight layer will be the ceiling and will be created by stripping back the existing plasterboard and replacing with OSB, joints and wall junction taped  then a service void and new plasterboard.  Other than a strip to plaster in the perimeter tape around the ceiling and making good where needed the existing plaster will be left untouched. So for the walls we will be relying on the existing plaster finish for airtightness. Or should we go with the external parge coat method?

The problem I have with this is the detail to join the ceiling airtightness layer  to the external face of the wall at the eaves will be nightmare and near impossible to get right on site. We could rely on the existing internal plaster and do the parge coat as an additional measure, but I don’t like the idea of two spate air tightness layers, as a wise man once said, you don’t want to fix a leaky bucket by putting it in another bucket with a hole!

Any thoughts appreciated

Ruairi

[Mark Siddall]

Hi Ruairi,
In our project we are going external so that we can avoid the very problems that you are encountering. Ai leakage behind internal stud partitions that abut the external walls is another one, and of course floor interfaces. Addressed internally they all require extensive stripping back and disruption. In my view it's really problematic and is best designed out. (I have a hunch that it's cheaper that way also.)

Sorry to be a doom sayer. If I have any more constructive feedback I'll let you know.

Mark

[Nick Grant]

Hi Ruairi

Normally agree with the bucket analogy but may be best option to parge externally but hope that the overlap between internal plater from ceiling down to first floor is enough to seal between. Would be different if was cavity wall or very open rubble. I think the real weakness if floor joists etc which are best dealt with from outside.

Any air-leak has to go through wall at first floor level and travel vertically through the mortar joints and pop out through perps in top course of bricks.

See what Alan thinks but I'd put money on this being OK.

Nick

[Ruairi]

Thanks for the responses guys.

Sounds like parge is the way to go, as stated, floor joist and partition junctions will always be weak points if we rely on the internal plaster. Still a bit concerned about the eaves detail, so I will work on this see what we can come up with.

R

[Mark Siddall]

New day new thought:

Depending upon access to the interfaces I recall that Dr Feist mentioned that gypsum expands when drying (rather than shrinks). Perhaps use a membrane where ever you can and then treat locally with poured gypsum. Reportedly worked well around timber.

Mark

[fostertom]

That could be a good one - how can we check that gypsum really does that? Tho I do recall seeing many hairlines where plaster has shrunk away from solid objects - or maybe only from timber, so it may be that the timber's shrunk away from the plaster. In that case, pouring around joist ends wd work only for old, fully-pre-shrunk timber. And what about slight joist-end movement due to floor deflection - that would force open a hairline in the plastercast.

[Mark Siddall]

Tom,
I did not say this was perfect. It is a remedial measure for a compromised situation. Hairline cracks may occur - but it's a vast improvement upon the existing condition. I guess that it needs to be >25mm thick to be robust. Has been done in PH retrofit schemes in Germany.

Mark

[Nick Grant]

My guess is that inner plaster to outer parge connection via bricks will be more airtight than floor joists ends could ever be made plus that sounds like a fiddly and time consuming job to do well.

 

[mike whitfield]

I also heard Dr Feist's expanding gypsum tip at the Oxford conference and tried it on a barn conversion, where existing oak tie beams passed through the new t/f wall we were lining it with. The gaps b/w the tie beam and the i-beam studs were about 300mm longx10mm wide. It seemed to give a tight joint across the 10mm, but shrank and cracked along the 300mm dimension. I think it has potential as a solution but needs a trial on your job. Other problems are that gypsum is very hygroscopic, and ours were exposed to the 'summer' for a few weeks, became very damp and didn't seem to dry out when we had the roof cover on, presumably because of the small surface area and deep poured joint. That wouldn't happen on the top of a wall though as long as you replaced the roof above straight away. Depending on the detail, if you can do a simple shuttering and pour say 15-20mm thick, it could be a cheap quick not too fiddly solution. On a 300mm thick wall, I think you might get about 4 lin m per bag at 18mm deep??
Mike

[Mark Siddall]

Nick,
Agreed but you still have the joint at eaves level - between external parge and the existing ceiling/air barrier - unless you decide to go up and over the trusses (which is of course ideal).

Mark

[Michael George]

That could be a good one - how can we check that gypsum really does that? Tho I do recall seeing many hairlines where plaster has shrunk away from solid objects - or maybe only from timber, so it may be that the timber's shrunk away from the plaster. In that case, pouring around joist ends wd work only for old, fully-pre-shrunk timber. And what about slight joist-end movement due to floor deflection - that would force open a hairline in the plastercast.

Re gypsum against timber. In more than 25yrs of plastering I have never known gypsum not shrink away from timber, or indeed anything attached to timber. I believe it is part of the reason why Cornice was developed [ie to hide the cracking between wall and ceiling] Also why we use architrayves around door frames

[Mark Elton]

Mark, as I understand your projects, you are doing both houses in a semi-detached pair which makes it easier to adjust the roof line, add insulation over the rafter etc. In our case, we have a semi where the other is in private ownership, making it necessary to retain the roof line. This highlights the big flaw in R4tF! Guess we can work on the eaves detail to take the air barrier to the o/s face of inner leaf/wall plate and rely on Nick's thoughts on difficuly air path through inner leaf.

[Mark Siddall]

Mike G,
Is it not that the timber shrinks as a result of changing moisture content?

Mark,
Yes the project that I am working on is somewhat dfferent in this respect. That's why after some reflection I suggested the gypsum option :-)

Very interesting to hear what Mike W has to say about trialing this idea. I think that in the the good Drs example a try was formed, this was lined with air barrier and the gypsum was poured in - local to the penetration it leaked into all the fine gaps in the timber.

Mark

[Michael George]

Mike G,
Is it not that the timber shrinks as a result of changing moisture content?

Mark

Yes, it is, but where there is direct contact between gypsum and timber I believe the process to be:
1. Wet gypsum applied
2. Moisture transfer from gypsum to timber
3. Timber swelling, though not visibly
4. Timber returns to previous MC, leaving hairline crack
Another example of this is where the top edge  of skirting can 'cup' if it hasn't been primed prior to plastering.

Of course seasonal expansion through both temperature and moisture content cause such minor cracking as well

[Mark Siddall]

Mike
I'm not disputing that there is a hairline crack but we've got from the width of a pencil down to hairline. That's a good improvement - though not perfect. The good thing about the gypsum was that it could leak into all the cracks in the old timber - as far as I know silicones would not achieve this as they are too gloopy.

Mark