Understanding insulation options and condensation implications

[Alan Ambrose]

Hi,

 

I'm trying to wrap my head around insulation options particularly re condensation. I think I kind of 'get it' re structural, installation & heat flow/temperature, but not not moisture.

 

I get that as the temperature decreases from inside to outside (generally in the UK anyway), for any particular humidity level, at some point we reach dew point and get condensation. And 'that's bad'.

 

But surely, as the temperature graphs must always fall off 'monotonically' i.e. always decreasing from inside to outside then we're going to hit the dew point at some position in the wall build-up?

 

Here's a commercial example. If I've understood their literature then the build-up looks like this:

 

 

and crunches out something like this:

 

 

- and the temperature and humidity graphs look like this:

 

 

That is, 'not so happy'. Any thoughts?

 

TIA, Alan

[ProDave]

And what does it look like if you move the polythene from in between the two different types of insulation to inside the PUR?

[Iceverge]

47 minutes ago, Alan Ambrose said:

Hi,

 

I'm trying to wrap my head around insulation options particularly re condensation. I think I kind of 'get it' re structural, installation & heat flow/temperature, but not not moisture.

 

I get that as the temperature decreases from inside to outside (generally in the UK anyway), for any particular humidity level, at some point we reach dew point and get condensation. And 'that's bad'.

 

But surely, as the temperature graphs must always fall off 'monotonically' i.e. always decreasing from inside to outside then we're going to hit the dew point at some position in the wall build-up?

 

Here's a commercial example. If I've understood their literature then the build-up looks like this:

 

 

and crunches out something like this:

 

 

- and the temperature and humidity graphs look like this:

 

 

That is, 'not so happy'. Any thoughts?

 

TIA, Alan

 

I've spent far too long playing with Ubakus. It's one of my favourite games (sad I know!) 

 

 

It's an excellent tool but not without it's short comings. Here are are a few:

 

If you put a layer of something like PIR boards it assumes that they form an immaculate seal from board to board. In reality unless you diligently tape the foil faces or use an immaculately sealed vapour barrier this isn't the case.  Have a look at this. 

 

 

The wall is entirely enclosed at both sides by a vapour barrier. It looks fine but is bad news in reality as the wall can't dry out. The program assumes that there's no failure points in the barriers and the internal construction was perfectly dry from the get go. Completely unrealistic. The above wall WILL have condensation and rot problems in real life. 

 

Similarly it assumes that all moisture is via diffusion (where moisture laden air seeps through materials) This is completely unrealistic too. Moisture gets through a structure by gaps and air currents. This is one of the main reasons that airtightness is so stringent in Passivhaus's as it is so important for building durability. 

 

 

Next have a look at this wall. 

 

 

It appears that it should be saturated inside however this isn't the case if a good job has been done with the airtightness layer. ( External sheathing in this case) 

 

 

 

This guy lives in a very cold area with much greater pressures for the "cold sheathing" problem than the UK would have. If you take care of the airtighness and don't trap moisture in a structure you won't have any issues. The cellulose really helps here. It's much more resistant to any airflow than mineral wool batts etc. 

 

 

 

 

Finally here's an example of an old fashioned TF with 2x4's on 16" centres with mineral wool in between and no vapour barriers. 

 

 

Again it looks like a disaster zone . In reality if it was in a location like Northern USA or Canada where -5 Deg and 80%RH is normal then it would condensate and rot out. However these conditions for prolonged periods are really unusual in the British Isles and lightly in a few weeks it'll be warmer again and the whole thing can dry out with no lasting damage. There's mountains of walls like this in perfect structural condition. 

 

 

 

 

 

It's a really tricky concept. 

 

The main principals are . 

 

1. Don't trap moisture in the structure. One Vapour BARRIER   max.

2. Poor Airtightness is how moisture gets into walls and roofs. 

3. If you can't guarantee the airtightness keep your vapour barrier inboard of the insulation to make sure it doesn't have any condensation or else build your wall entirely out of vapour open materials. 

4. Having the outer part of your wall below the dew point is ok for short periods so long as it can dry effectively again. 

 

 

 

 

 

[Iceverge]

As an aside on your example. 

 

1. You can select "roof" from the drop down wall menu rather than "wall"

2. You have polythene in the diagram but polyethylene in the example. These will have different permeability. 

3. PUR insulation is different from PIR.

 

Happy insulating!!!

[Alan Ambrose]

>>> And what does it look like if you move the polythene from in between the two different types of insulation to inside the PUR?

 

Ah thanks, spotted (one of my) mistake(s)  - their section drawing does show it where you suggest. That makes it look a bit better with only the OSB apparently having the main condensation risk...(left graph)

 

I guess if I take the min temperature as +1.5C (which is the actual mean minimum from historical data in our area) instead of -5C then that makes it look even better (right graph). I suppose I'm assuming it'll dry out OK if the temperature occasionally dips below the usual minimum ..... this really does appear quite a complex subject

 

[Alan Ambrose]

@Iceverge - thanks for taking the time to put up those examples - I think I'm going to have to carefully work through your examples to make sure I fully understand all your points

 

I'm sure my brain was faster at some point than it is now ...

 

[SimonD]

2 hours ago, Iceverge said:

or else build your wall entirely out of vapour open materials. 

 

Or even better make the wall buildup hygroscopic and vapour permeable. I personally think there is still a lot left to learn regarding building physics and moisture movement in modern building practise, especially regarding hygroscopicity at its role in effective moisture buffering.

[Iceverge]

1 minute ago, SimonD said:

 

Or even better make the wall buildup hygroscopic and vapour permeable. I personally think there is still a lot left to learn regarding building physics and moisture movement in modern building practise, especially regarding hygroscopicity at its role in effective moisture buffering.

 

True, you could dedicate a lifetime of study to it. The basics aren't rocket science however. 

 

All you need to do is avoid a build-up of moisture. 

 

How you do this is the next question. 

 

@Alan Ambrose I've put a suggestion on this on March 7th. I'll stick it through Ubakus now. What are your target U values and I'll try to design a cheap and buildable solution. 

 

[Iceverge]

 

 

All very buildable. No exotic materials. I didn't bother drawing the external shiplap and battens. You could replace the inner Airtight membrane for another layer of OSB and tape it if for airtightness if you wanted.

 

 

 

 

Edited November 21, 2023 by Iceverge

[Alan Ambrose]

Hi @Iceverge - I don't get the ref to March 7th and I would feel guilty about taking up a bunch of your time. But ... if you have any quick-and-dirty hints - I'm looking for u-value of ~0.1, to hang on the outside of a structural oak frame. Preferably not 0.5m thick

[Alan Ambrose]

Totally cool, thanks so much - now I have a bunch of home work to do

[Iceverge]

2 minutes ago, Alan Ambrose said:

I'm looking for u-value of ~0.1, to hang on the outside of a structural oak frame. Preferably not 0.5m thick

 

Cool, 0.1W/m2K takes a certain amount of thickness and once you get much below about 0.2W/m2K (Educated guess, happy to be corrected) thermal bridges become really far more relevant. In other words just throwing insulation at a notional wall build-up won't make any real difference unless you get your window and door junctions sorted etc. By far the easiest way to do this is some is some kind of external insulation over the frame. However as you are planning on timber siding hung from the exterior wall,  there's a limit to how thick you can go as long screws to tie battens back to the frame are not cheap either and may be result in the siding cantilevering too far from the support point. 

 

Something like @IanRs wall might be an option. 

 

I-Joists full filled with cellulose and an Egger MDF layer outboard taped for airtightness. I think he has maybe the best airtightness score of anyone on here. As you need to hang plasterboard between the Oak frame you'd still need some internal timbers to allow for this as the I joists wouldn't line up correctly. ( Ubakus can't draw these horizontally so you'll have to imagine it) 

 

[Iceverge]

Pro's:

 

All you insulation is in the one place with no waste. 

Egger DDF offers a chance to externally insulate window frames to some extent. 

The web of the I joist offers a good chance to take screws for external battens. 

Easy Airtightness details boosted by cellulose would make a great score possible with some care. 

Plenty of area to run wires etc inside the 44x95mm horizontal battens. 

Tremendous decrement delay. (Phase shift) 

Vapour open construction = very safe long term. 

All you Oak frame is inside the insulation zone making it completely temperature stable and unlightly to move at all

Flexible insulation etc so any small movements make no difference.

Floating Plasterboard not connected to the lovely Oak frame. making it simple to fit and unlightly to crack. 

 

Con's:

You're moving into slightly more specialised and dearer materials  like the Egger DDF, if you run a board short you won't be able to buy one locally, you'll have to wait for a delivery. 

Similar argument for the I joists. They are dearer than cut lumber. 

Cellulose isn't as cheap as some insulants on a £/U value basis on the face of it. However once you take into the account the cost of fitting and zero waste I think it's good value. We have 450mm in our attic.

You'd need to somehow add rafter extensions to the I joists if you want to have an overhang of the roof on your walls. Defiantly recommended if you want have timber siding.  

 

 

 

 

Edited November 21, 2023 by Iceverge

[Iceverge]

 

 

Here's a quick 3d View. 

 

From the outside. 

 

30mm egger DDF

360mm i joists @600mm cc filled with densepack cellulose

44x95mm timbers horizontally also filled with cellulose to hang plasterboard on. 

15mm plasterboard 

100x20 vertical strapping on 250x250mm oak uprights to allow plasterboard to slip being oak. 

 

 

 

[Iceverge]

 

If you were able to guarantee an excellent job taping the external OSB for airtightness I would have no issue with the above. 

 

Replacing the 360mm I joists with 220x44 mm timbers and the  Egger DDF with 11mm OSB would be much much cheaper and simple to source locally. 

 

Also rotating the 95x44mm horizontals would give a little extra insulation thickness for a very respectable U value. 

[IanR]

2 hours ago, Iceverge said:

I-Joists full filled with cellulose and an Egger MDF layer outboard taped for airtightness. I think he has maybe the best airtightness score of anyone on here. As you need to hang plasterboard between the Oak frame you'd still need some internal timbers to allow for this as the I joists wouldn't line up correctly. ( Ubakus can't draw these horizontally so you'll have to imagine it) 

 

Interestingly, Oakwrights (where the image in the first post came from), have almost this solution for their PassivHaus option. They just don't include any insulation in the service void, and use Panelvent Sheathing instead of Egger.

 

 

I have a friend with a beautiful, Oakwirghts' Framed home. It really is lovely. It has been a labour of love for him though to keep up with all the movement of the frame. Every year he goes around filling, caulking and redecorating. He's about 11 or 12 years in now, and it's still moving. At times there have been gaps between the Oak frame and the plasterboard that you could get you hand through.

 

For the PassivHaus option, I assume the Oak Frame is independent from the I-Joist Frame, which seems quite an expensive option.

 

@Alan Ambrose, What's your expectation of the Oak Frame? Will it be a green oak frame and will you be using someone like Oakwrights? Do you want the oak frame to be visible externally, at all? Do you want frame to be structural, or are you happy for it to just appear to be so. 

 

To combine a green oak frame with Passivhaus type aspirations is likely to lead to some redundancy, and therefore a higher cost, and maybe some compromises.

 

Edited November 21, 2023 by IanR

[Iceverge]

I've reinvented the wheel again haven't I!

 

That pic above very much looks like the oak frame is just for show. The I beam frame is raised first, then plasterboarded and then the oak frame. It'd be a durablen if expensive solution though. 

 

If they're done it the other way around it'd be a nightmare airtightening the propassive behind the oak frame and the plasterboard would definitely crack too if screwed between the oak and service batten. 

 

 

Pondering this I wonder would something like a hempcrete wall be an option. Its self supporting but not structural so would tie in well with an oak frame. You'd need 700mm walls to get near 0.1W/m2K though. 

 

 

 

 

[GaryChaplin]

12 hours ago, IanR said:

He's about 11 or 12 years in now, and it's still moving.

Blimey, they must be big timbers! As a rule of thumb, the oak takes about 1 year per inch thickness to dry. So an 8 inch (200mm) sq beam takes 8 years to dry. However, in a heated dry environment you can half that. 

 

Caulking gaps whilst the frame is drying is a bad idea, it will just look like a mess. Much better off waiting for the frame to dry and re-plaster. The ultimate solution is to put the green oak frame together and leave it for about 10 years, then finish the job!

[SimonD]

8 hours ago, Iceverge said:

Pondering this I wonder would something like a hempcrete wall be an option.

 

I'm not so sure. I had a good look round the WISE building at CAT on a visit a good few years ago. Lots of cracks/gaps visible at various intersections at the time. I'd be curious to see how it performs now for airtightness.

[Kelvin]

I have friends with a timber oak building. Same issue with gaps etc except they’ve not bothered filling the gaps. It’s 8 years old and looks 60 years with all the wonky gaps. Oddly though you tend to look beyond that and admire the craftsmanship and the timber. 

[Alan Ambrose]

Goodness, the BH hivemind never ceases to impress me - I'm still chewing through all the detail in the above few messages. @Iceverge - apologies if I used up any of your time unnecessarily - I didn't spot the other Oakwrights 'Natural' build-up.

 

To answer IanR's questions:

 

>>> What's your expectation of the Oak Frame? Will it be a green oak frame and will you be using someone like Oakwrights?

 

Yeah, I'm sure it'll move a bit - I'm not fussed though and actually like the living nature - most of our 'engineered' materials are so dead. We live in an old barn conversion right now - the frame (prob 250 years old) is all over the place and obviously made with whatever oak sticks the farmer had around. And it's all the more beautiful for that.

 

>>> Do you want the oak frame to be visible externally, at all?

 

Maybe just through the windows.

 

>>> Do you want frame to be structural, or are you happy for it to just appear to be so. 

 

Yeah structural - I like the idea of hanging the rest of the build up on the frame. I'm also attracted to steel frame for the same reason - probably too much time with Meccano when I was a kid. I can see with our existing barn though that it does allow the frame to be used again i.e. 'recycled' quite easily by tearing off and then replacing the 'cladding' - maybe in a couple of hundred years time.

 

 

Edited November 22, 2023 by Alan Ambrose

[SimonD]

24 minutes ago, Alan Ambrose said:

Yeah structural - I like the idea of hanging the rest of the build up on the frame.

 

You could look at Glulam - Buckland Timber, who I can recommend, do glulam in Oak as well as a number of other timber species. This could be advantageous for a number of reasons. https://www.bucklandtimber.co.uk/

[Iceverge]

1 hour ago, Alan Ambrose said:

@Iceverge - apologies if I used up any of your time unnecessarily

 

No problem. Its an interesting topic. 

 

It's annoying to effectively double up on structure, but that seems to be difficult to avoid if you want an oak frame that performs well. 

 

The expense also annoys me as the modern equivalent of a steel portal building clad with insulated metal panels is one of the most economical build methods. 

 

 

[Alan Ambrose]

Wow thanks - impressive what you can do with glulam.

 

Thanks again @Iceverge - you've educated us a lot and I feel I'm getting to grips with the options / challenges now. I'm guessing you're an SE or other engineer?

 

For everyone's amusement, I'll post up here my nominally selected stack-up when I get to that point. Here's where I am thus far (these are all options without any insulation in the service void) ...

 

[Alan Ambrose]

>>> It's annoying to effectively double up on structure, but that seems to be difficult to avoid if you want an oak frame that performs well. 

 

Yeah, in a structural sense, there is a slight bananas aspect to it, .

 

>>> The expense also annoys me as the modern equivalent of a steel portal building clad with insulated metal panels is one of the most economical build methods.

 

Point taken.