Designing walls that are not vapour permeable - A good idea?

I have been having a discussion on a LinkedIn Passive House forum regarding the choices one can make in regards to insulation and the effects of these choices.

http://goo.gl/1vGTyI


The poster was asking for experiences within the building community with Wood Fiber vs. Cellulose insulation and I suggested that neither may be desirable depending on your climate conditions and instead suggested continuous exterior mineral wool fibre insulation.  This then morphed the conversation towards what constitutes a durable high performance wall.

I posted my thoughts on the perfect wall (which just happens to match my walls in my upcoming build) and other who are builders of PassivHaus (PH) structures posted their perfect wall details.  This led to a discusion about the merrits of designing a wall that is vapour open to the low pressure side, where one of the posters stated:

"I've been indoctrinated with the Bau-biology "Breathable Wall" idea with nearly 10 years now and spent many years preaching that gospel. But then I found the Spokane and Tsong studies where they opened the walls of 250 houses, that were built wrong in terms of 5:1 breathability but no decay was found.  The walls had no membranes, no decay was found, its the same for SIPs houses, ICF houses and most other construction methods, the walls don't breathe as per the 5:1 rule and the houses aren't falling down."

The two studies can be found here:
1) http://www.viking-house.ie/downloads/Tsong79.pdf
2) http://www.viking-house.ie/downloads/Spokane.pdf

I read the two studies the poster provided and was somewhat shocked at the jump in logic that is represented by the statement that we do not need to make walls permeable and that impermeable walls will not rot. This is such an important subject, I though I would reproduce my comments here to a larger readership.

The Tsong study is discussing the lack of VB, and not a wall that is vapour tight. A wall that does not have a VB is by definition VERY vapour open and in fact most of the assemblies studied were quite vapour open (poorly insulated wood frames).  It is also important to note that the study occurs in 1979 and the levels of insulation discussed are far below what we are talking about in today's high performance homes (the study does not state the R value but we are talking about poorly filled 2x4 walls, so probably an effective average of below R7). Therefore these walls all had a lot more drying potential due to thermal bridging than high performance homes of today and certainly a lot less drying potential compared to a PH. It is also important to note that these houses had an average ACH50 of 16.2, which is more than enough to also help dry the assembly when it was experiencing very high RH levels. I have been unable to locate the permeability of Urea Formaldehyde insulation so do not know how permeable those walls were if detailed perfectly – but per the study, these foam walls had a lot of air leakage due to foam shrinking and cracking. The average foam shrinkage was 8% and the report states that as a result of the shrinkage of this foam, there was a 70% increase in heat loss (heat loss dries walls, so even these walls could dry easily).

It should also be noted that areas of high moisture content were found at many locations on these homes where bulk water entry was occurring (in other words control layers regularly fail and you should design your assemblies for such to the extent possible).

What I do love about this study is their remark at how the mineral wool insulation had an ‘extremely low average moisture content’ when compared to the other insulation (in no case was the moisture content of the mineral wool above 2%). The study went on to say this is “probably attributable to the fact that mineral wool is not hygroscopic, whereas the cellulose and U-F foam both tend to retain moisture”. Go ROXUL!

I then went on to say that relying on this dated research to state that a wall should not be vapour permeable to the low pressure side is grossly flawed in my view, does not come close to lining up with the current recommendations of the building science community and their experience in repairing failed structures, and in my view also miss-interprets the studies results and compares conditions that are grossly miss-aligned with the high performance structures we are building today.

All this study can really claim is that there was no significant moisture damage associated with diffusion observed on any of the homes that generally had no or minimal insulation and high levels of thermal bridging. And as we know today, this results in assemblies with built in drying safety factors. The study was also was clear to specify that these results could not be related to colder climates.


The second study was by the same author but took place in a colder climate.  The same arguments above apply.  It should also be noted that colder climates generally have less problems with moisture damage to wall assemblies than milder climates.  In climates with cold winters and hot summers, the moisture typically exists as frost all winter and then quickly dries out in the late spring as the temps rise. In a location like the Pacific North West (3000 DD), moisture will stay in liquid form for months at a time as is able to cause a LOT more damage as the moulds take hold and flourish.

Can you build vapour tight assemblies that work in the Pacific North West?  Yes, but you then need to sweat the details.

An ICF IS a wall that works.  It is often quite vapour tight but because there is no air movement through it at any point in the structure, there is generally no opportunity for condensation to occur (I have heard of isolated events where condensation has occurred between the foam ICF and concrete core leading to mould build-up).  The typical foam materials of the ICF are also highly resistant to vapour diffusion, all but eliminating that risk as well. From a building science standpoint a typical foam ICF structure makes a lot of sense, but where it fails in my view is the very high embodied energy that it represents both in terms of the volume of concrete used in these homes and the foam used in the typical ICF blocks.

As far as SIP construction (structural insulated panels), which are typically fabricated with OSB sandwiched on each side of a foam block, I personally feel that the jury is still out.  There is a multitude of reports of SIPs failures across North America and once again, this style of construction represents a high embodied energy.

For me, I will stay true to my stick frame, plywood sheathed structure wrapped in a nice continuous warm blanket of highly vapour permeable and fire/rodent/bug proof Roxul mineral wool insulation thank you very much.