What is Ubakus really doing? And what is that 2/3 - 1/3 rule?

[Garald]

I've read Ubakus really uses a very simple model to predict condensation. So, what is it, and where can I read about it? What do more sophisticated models (such as the one used by WUFI, which I can't use, as I don't have Windows - just failed to run it using WINE) really do?

 

TL;DR: I would like to install outside insulation on a wall that is already insulated on the inside. I want to actually understand (the physics behind) how to make choices so as to prevent condensation - I can't just follow rules of thumb such as the "1/3-2/3 rule" and "at least 5 times as much resistance to water vapor on the inside as on the other" as they can't really be simultaneously obeyed in this sort of situation.

[SteamyTea]

The physics is quite simple.

Air is a mixture of gases. Nitrogen, oxygen, water vapour, argon and the other traces gases, carbon dioxide, methane, nitrous oxide.

 

Nitrogen and oxygen are fairly thermally stable in the earth atmospheric temperature range.

Water vapour (about 0.25%) is anything but stable. It can be a solid, a liquid or a gas. Sometimes all three at the same time (triple point).

Temperature is not the only thing that makes it change state, air pressure can as well.

This is why when you look at some more sophisticated RH and AH models they ask for air pressure.

 

Water is also strange molecule. It has different thermal properties and different temperatures, and to make matters worse, expands, rather than contracts, when between 277K and 273K, with the latter temperature being a phase change temperature.

 

Then there is the way that it likes a nucleus point to change phase on. Mineral wool gives it billions of these, so like to condense and freeze in it.

 

As water changes state, it releases a lot of energy, which can warm surrounding gases and solids. This prolongs the phase change time. 

 

It really is a messy business.

 

But there is another way to model it.

Statistically.

Build an array of phase change data points that correspond to the local RH, temperature and maybe air pressure.

Then look at the probabilities.

You will generally find that the high risk times i.e. condensation does not happen very often, or for very long.

 

It is also worth remembering that a wall has surface temperatures slightly lower than internal air temperature on the inside, and often much higher than outside air temperature on the outside where it is heated by the sun.

 

Our old mate @Ed Davies wrote and interesting bit about humidity on his website.

https://edavies.me.uk/2017/03/vapour/

[Alan Ambrose]

From memory doesn’t ubakus allow you to choose from a bunch of different moisture models? I’m guessing that a typical moisture analysis usually isn’t so marginal that the choice of moisture model means much anyway.

[MikeSharp01]

5 hours ago, SteamyTea said:

 

Our old mate @Ed Davies wrote and interesting bit about humidity on his website.

Anybody know what has become of Ed? Not seen here for some time and his last blog post is also some time back. Hope all is well.

[SteamyTea]

On 23/06/2024 at 22:04, MikeSharp01 said:

Anybody know what has become of Ed

https://edavies.me.uk/2023/10/abandoned/

 

Seems he has not posted at the other place since October 2022.

Edited June 26 by SteamyTea

[Alan Ambrose]

That's sad, it looks like it was going well.

 

Re Ubakus: yes, here are the moisture models. The OP seems to have vanished, but I would encourage him to try the model options and see what difference it makes.

 

 

[Garald]

51 minutes ago, Alan Ambrose said:

That's sad, it looks like it was going well.

 

Re Ubakus: yes, here are the moisture models. The OP seems to have vanished, but I would encourage him to try the model options and see what difference it makes.

 

 

I haven't vanished!

 

At any rate, I take the DIN this-and-that models are very simple 1-D models one can do with pencil and paper?

[SteamyTea]

On 26/06/2024 at 21:40, Garald said:

I take the DIN this-and-that models are very simple 1-D models one can do with pencil and paper

We can do those simple models, and probably use a 'solver' to make it a 2 or 3D model, but not many people can.

[Alan Ambrose]

I’m guessing that they are 1D differential equations - probably more suited to simple code that pencil & paper.

 

Did you test out what difference the various moisture models made to your stack-up or not?

 

 

 

[MikeSharp01]

1 hour ago, Alan Ambrose said:

I’m guessing that they are 1D differential equations - probably more suited to simple code that pencil & paper.

I am guessing that they would not have developed the Wufi system if they could have done it with a spreadsheet. This field is littered with no fun hard sums the most difficult of which is calculating value when all you know is the cost 🙄🤣.

[Garald]

4 hours ago, SteamyTea said:

We can do those simple models, and probably use a 'solver' to make it a 2 or 3D model, but not many people can.

Well, can you tell me where to look up those models (and the physical reasoning behind them)?

[SteamyTea]

4 hours ago, Garald said:

Well, can you tell me where to look up those models (and the physical reasoning behind them)?

Matlab/Freemat could do it.

 

I really would not get too deep into the physics, it is the results that are important.

But as I said earlier, a statistical model is easier and good enough.

[Garald]

2 hours ago, SteamyTea said:

Matlab/Freemat could do it.

 

I really would not get too deep into the physics, it is the results that are important.

But as I said earlier, a statistical model is easier and good enough.

 

If you wish, but where do I read up about these models?

[Alan Ambrose]

Google the DIN standards maybe? Easily available for sale but only in German (like a lot of PHPP detail). Maybe there’s an English language explanation somewhere on the web.

[MikeSharp01]

2 hours ago, Garald said:

If you wish, but where do I read up about these models?

There is quite a lot in the academic press around the Wufi stuff, probably also Ubakus but I have not looked, Wufi has been around sine the 1990s and was IIRC started as an academic exercise- once they knew what they had, something that worked, was useful and that very few people understood in any detail, they went quiet on it a little because they saw the commercial potential and went that way. None the less although it is possible to get to grips with it I suspect that for most people it won't repay the effort either intellectually or financially which is why we paid to have ours done.

[Garald]

1 hour ago, MikeSharp01 said:

There is quite a lot in the academic press around the Wufi stuff, probably also Ubakus but I have not looked, Wufi has been around sine the 1990s and was IIRC started as an academic exercise- once they knew what they had, something that worked, was useful and that very few people understood in any detail, they went quiet on it a little because they saw the commercial potential and went that way. None the less although it is possible to get to grips with it I suspect that for most people it won't repay the effort either intellectually or financially which is why we paid to have ours done.

 

Well, I hope that this is the one and only renovation I will do in my life, so any knowledge I acquire is useless to me *except* for academic knowledge (which is a nice way to review undergraduate physics, learn what our friends in finite element analysis do after all, etc.).

[Iceverge]

I've been thinking about this for about 8 years now since starting the house process. 

 

Any model is naturally limited in its output because there is only a finite amount of inputs. Ubakus for instance will throw up moisture trapped within a wall for some buildups, however that is only when the outside air temp is at the default of -5.  Whilst most of us might get that occasionally it's almost never long enough to cause an issue. If you live in Siberia it might be a different story. 

 

This issue is easily solved with some analysis of the most lightly worst case weather scenarios based on historic data and future predictions. 

 

However what is impossible to predict is the exact amount of showering or stir-frying the occupants do. Now hot they heat their house, how long the kids leave the door open. Their choice of paint for the walls, or even how many coats of it. 

 

Any model which will assure you of this is wildly pessimistic. Any guaranteed moisture safe construction is hugely conservative  with its vapour management and the as built thermal grading. Take an ewi wall and a cold vented attic as examples. 

 

 

 

 

 

 

 

 

 

Edited June 29 by Iceverge

[Iceverge]

It took me 6 years to have my Eureka moment. 

 

The only principle that matters is:

 

The average amount of DRYING must exceed the average amount of WETTING. 

 

This is the only thing that matters. If you heat and ventilate any space sufficiently then there will be no problem with moisture buildup regardless of insulation and airtighness layers etc. 

 

Take an aircraft for example, the impermeable skin is outside the minimal insulation, often at sub zero temperatures but there's no issue with moisture despite hundreds of people breathing it out. That is because they are ventilated and dehumidified to within an inch of their lives by the air-conditioning. 

 

A well built thermal bridge free house with a vapour impermeable barrier inboard may still suffer moisture problems with excessive moisture generation and poor ventilation. 

 

It is a highly dynamic situation and in reality all you can do is make a best guess with a model then suck it and see. If you find yourself at the wrong side of things then you need to add DRYING and reduce WETTING. 

 

 

 

 

 

[Iceverge]

Do this by reducing internal humidity, more heating and ventilation. Less drying of clothes, diligence with extractor fans and cooker hobs. 

 

Adding more insulation outboard if possible like EWI. Eliminating multiple layers of highly impermeable materials into the wall to facilitate frying both onwards and outwards. 

 

Improving airtighness to prevent air cracks taking moist air into the wall. 

 

 

[SteamyTea]

45 minutes ago, Iceverge said:

The average amount of DRYING must exceed the average amount of WETTING

While I agree with the principle, it is a skewed distribution because there is more chance of mould forming at higher temperatures, and there is difference in phase change temperatures and the associated SHC for different states.

[Iceverge]

 

On 29/06/2024 at 14:07, SteamyTea said:

While I agree with the principle, it is a skewed distribution because there is more chance of mould forming at higher temperatures, and there is difference in phase change temperatures and the associated SHC for different states.

 

The dirty secret that is little talked about is the effect airtighess has on moisture related problems. 

 

Do a proper sealing job with one layer of low vapour permanently anywhere in your wall or roof,then stick insulation either or both sides of it, and problems just don't occur. 

 

Look at the houses in the US in mega cold climates with an external taped layer of OSB that survive just fine. 

 

Meanwhile badly constructed dry lined houses here end up as "fungus in your wall nirvana" despite looking fine in a model. The point nobody considered was the crap air sealing that is the norm.

 

Edited July 1 by Iceverge

[SteamyTea]

36 minutes ago, Iceverge said:

Look at the houses in the US in mega cold climates

The USA is generally south of us, and the cold places are at high altitude, with low humidity.

It is the warm, southern states that has the VLC on the outside, by default, because the houses are colder, and dryer, on the inside because of the air conditioning.

[Iceverge]

True that, Pacific NW gets plenty rain i'd have thought though. 

 

I must look up the design codes. God , what has my life become.