Double Vapour Retarders are NOT Fine by Me!

Far too often, a poster on LinkedIn makes comments that defy good building science.  As I am often busy, I try to bite my tongue and just move on, but often the posts push too many of my buttons and I find myself in the position where I MUST comment or commit hari kari!

A resent post titled "Double Vapor Retarders are Fine by Me!" was one such post that pushed me to reply:

I find it interesting that often when I see what (in my view) is bad building science, the individual in the conversation is almost always in the business of selling or pushing foam. The facts are that Physics has not changed – EVER- and the rules are not being ‘smashed’, only disregarded – often to the building owners peril.

Re OP - I believe this is propagating bad language and bad science and (wish) to review my view of the basics.

VB or vapour control layers are to address vapour movement by diffusion. The requirement for the ‘tightness’ of this control layer is based on the vapour gradient across the assembly and location is based on the direction of the pressure. The VB should always be on the high pressure side, so generally inside in heating climates, outside in most cooling climates, and carefully designed and managed in the rest of the climates. Diffusion is a very small vapour driver, and while it cannot be ignored, having a pretty good VB is more than adequate. You do not need to sweat the details on a VB and do not need to worry about holes and untapped seams. A 90% effective or even 80% effective VB is going to be just fine.

An AB on the other hand is critical, and even small holes in AB’s can move large amounts of moisture by means of convection. The AB can pretty much be anywhere in the assembly and many (myself included) like to detail this layer on the exterior of the sheathing, where the number of penetrations are reduced and much easier to detail. When on the exterior, the one issue to address from a thermal performance point of view, is to reduce convection currents within the stud bays by use of a denser insulation.

Can you build a wall assembly with two VB’s? Yes in theory IF you have a perfect AB always. But in practice this is foolish, in my view, and will come back to bite you a significant number of times, because AB’s are almost never perfect in the field and or do not stay so for long even if they start out perfect. So conventional wisdom, based on decades of experience by many, pretty much always recommend that the assembly is generally vapour open from the VB control layer towards the low pressure side of the assembly. Careful design using modeling may show that you get away with a barrier on one side high side and a retarder on the low pressure side, but (in) my opinion, this is only going to consistently work in dryer regions where the load is low anyway. 

The reason a freezer works is because the two metal shells on each side of the foam are perfectly sealed as air barriers and the manufactures go to great lengths to ensure this during assembly with both sealants and gaskets.

The OP mentions “any mistakes are easily corrected by the air leaks”. Really? Are you promoting an ineffective AB? Are you promoting air movement that can move HUGE volumes of moisture into and through a wall??

The only time a dew-point does not exist, is when the temp and humidity of the air on both sides of an assembly never reach each other’s dew-point. Obviously, this is extremely rare. You can however negate a dew-point in various fashions. A) You can remove all of the air in and through an assembly so that there is no moisture to condense. (easy to do in a manufactured fridge – hard in a site built structure) B) You can design your assembly so that your thermal control layer is mostly/all on the low pressure side of the VB and WRB control layers (so that the condensing surface never reaches the dew-point temperature).

The spray foam crowd claim that filling stud bays with foam meets strategy 1. But I have yet to see a spray foam stud cavity where the foam has not pulled away to some degree from the studs (ore probably has been compressed away from the studs as the studs expand and contract into a material with no elasticity). SO, in these circumstances, the ‘air tightness’ has been lost and you better have a Plan B.


In my Linked-IN Post I also commented:

• Both OSB and Plywood are a Vapour retader in dry conditions, but only plywood opens up to over 6 perms in a wet-cup environment, that for instance all sheathing sees in my region. 
• My vapour diffusion holes question was a bit of a trap. Only way they work is when you have air movement through them, which is obviously something you do not want happening. I did a lot or research before posting on this subject http://goo.gl/0SAiSJ

Of course, I should have also added a third method to my post for negating a dew-point, and that is to ensure that all materials down stream of the high pressure side and VB control layer are vapour open enough to allow drying to the low vapour pressure side. 

It really frustrates me, that in order to sell foam (spray or rigid) to the building industry, manufacturers, vendors, and installers all try to twist the science to meet their objectives.  While this will help them sell their product or service, it will often leave the building owners with assemblies containing much higher risk for condensation, rot, and mould. 

I feel it is up to the rest of us to cry foul when we come across these inaccuracies and try to provide some protection to potential victims of bad science.