June 17, 2013

An Interior Air Barrier Does It Better

Inside-out approach to high performance

Inside-out approach.  aka, backward.

As more buildings strive to achieve airtightness, a common approach we too often see places the air control layer outboard of the insulation layer – often with OSB or plywood – to the exclusion of an inboard air control layer.  On the contrary, if you only have one airtight layer, this is the wrong side – it is a high-performance building turned inside-out.

As airtightness requirements and goals have gotten more stringent:  1.5ACH50, 0.60ACH50 toward 0.30ACH50 and below, designers and builders, finding themselves on the frontiers of airtightness, have gone with what seems to be the most straightforward path to passing the blower door test:  a continuous exterior shell of OSB or plywood.  But this allows conditioned air to readily enter the insulation layer by convective driven air-leaks and diffusion. It also puts a vapor retarder outboard of the insulation and consequently doesn’t prevent warm humid air from reaching cold components, increasing the condensation potential, while the exterior retarder unhelpfully restricts the outward drying potential.*

Inside-is-the-right-side placement of OSB for high performance

Inside-is-the-right-side www.greengenerationbuilding.com

Instead, let’s think about what a high-performance assembly generally wants to be and not just what appears to be the simplest route to passing the blower door test.  To do so we need to acknowledge that, generally speaking, in cold/mixed climates, the exterior wants to be relatively vapor open. The exterior doesn’t need this to achieve Passive House airtightness (though tighter on exterior is better) – it just needs to be what we call windtight, to avoid windwashing degradation of the insulating layer.  A better assembly is possible if we concentrate on how to combine, on the interior, the airtight layer and vapor control.

The specifics of why an interior air barrier make more sense are these:

Interior air barrier - www.ecocor.us - photo credit: Naomi Beal

Interior air barrier – www.ecocor.us – photo credit: Naomi Beal

  1. It prevents the conditioned air from entering the insulation layer – keeping the conditioned air within the conditioned space.
  2. It provides better protection against condensation risk – keeping interior humid air away from cold components. (After bulk water intrusion, air leaks and convective movement are the enclosures’ biggest liabilities.)
  3. It places the components of this most important control layer in a climate controlled environment – protected from temperature extremes – and ensures maximum longevity.
  4. Leaks are more readily found and easier to repair – as you can typically stand directly next to the air barrier and feel the leaks during a blowerdoor test.
  5. The air control layer can double as a vapor retarding layer – where you want it: inboard of the insulation.

Consequently a mixed/cold climate enclosure from outside to inside generally wants to be:

  1. a protective back-vented rainscreen and vented roof
  2. a vapor-open, windtight, water control layer
  3. fiberous insulation
  4. a vapor variable retarding airtight layer
  5. a protective service cavity with interior finish
Interior airbarrier that is verifiable and repairable.

Interior membrane airbarrier that is verifiable and repairable.

This inboard control layer is best composed of OSB or plywood sheathing (taped with TESCON Vana), or vapor variable membranes like INTELLO Plus or DB+. The fiberous insulation is best as dense-pack, but equal performance can be achieved with batts given the below configurations.

High Performance Assembly with airtight/vapor variable sheathing inboard of insulation layer

High Performance Assembly with airtight/vapor variable sheathing inboard of  primary insulation layer

In masonry buildings with interior insulation it is best to make the masonry windtight but to put the airtight layer inboard of the insulation.

Masonry retrofit with inboard airtightness.

Masonry retrofit with inboard airtightness.

If using ZIP 1/2″ OSB as the outboard wind-tight layer, use INTELLO as the interior air barrier to ensure the assembly is 5x diffusion tighter on the inside than on the outside in winter, yet allows inward drying in the summer/AC season.  See also this blogpost about vented rainscreens.

*OSB dry cup permeability ranges 0.5-0.8, depending on thickness.  Plywood ranges 0.7-1.4, depending on wood type and thickness.  Sources: ASHRAE fundamentals, ASTM E96 dry cup.

Related Blog Posts:


, , , , , , , ,

Tweet about this on TwitterShare on FacebookShare on LinkedInPin on PinterestEmail to someone

18 Responses to An Interior Air Barrier Does It Better

  1. Robert Riversong June 18, 2013 at 4:03 pm #

    As one of the pioneers of interior air barriers, breatheable envelopes, and the 1:5 inside-to-out vapor permeability ratio in cold climate construction, I appreciate your bias in that direction.

    But I question the use of potentially toxic (formaldehyde, a primary chemical sensitizer) materials like OSB adjacent to the conditioned space, the need for a rainscreen on any but the most exposed houses, and the additional materials needed to create both exterior and interior secondary cavities, as well as the wisdom of extreme air-sealing (such as the PH standard of 0.6 ACH50).

    An unintended consequence of over-tightening new homes is that the 2013 ASHRAE 62.2 standard now eliminates the credit for natural infiltration (2 cfm/100 sf surface area) and has nearly doubled the required volume of mechanical ventilation, thereby reducing energy savings and increasing operating costs.

    • foursevenfive July 3, 2013 at 10:17 am #


      I’m responding again to your post as recent technical difficulties seem to have wiped-out Floris’ previous reply.

      We definitely look to your work as a pioneer – it is inspiring and informative. Fair point on potential toxicity – if someone is concerned we happily offer non-toxic options in INTELLO and very green DB+ membrane.

      We are certainly proponents of the tighter the better. However, we would not put the responsibility of the 2013 62.2 fiasco on airtight enclosures. We are in agreement with Joe Lstiburek on this, who says “ignore it” – and that ventilation rates for typical homes should conform instead to 2003 62.2, but without the natural ventilation credit.


      • Robert Riversong July 3, 2013 at 12:52 pm #

        I don’t see the 2013 62.2 change as a “fiasco” but rather as a responsible emendment because of the decreased natural ventilation in extremely tight new homes.

        Without scientific validation, using the old mechanical ventilation standards with super-tight homes may not provide sufficient air exchange to eliminate interior moisture and other normal pollutants from furnishings and occupancy behavior.

        Both interior air quality and moisture management (which are somewhat connected) should take precedence over that last BTU of energy conservation.

        • foursevenfive July 3, 2013 at 1:03 pm #

          Scientific validation? As we understand it, and as Joe Lstiburek is described as saying on GBA (june 28), the original 62.2 rates were unjustified and the revision makes an unjustified situation worse. We disagree with Joe on use of foam, but think he makes sense on ventilation rates.

          • Robert Riversong July 3, 2013 at 1:56 pm #

            Actually, the original ASHRAE ventilation recommendations were as much as 20 cfm per person to deal with body odors, but they settled on 15 cfm/person and then halved that with the 62.2 standard to 7.5 cmf/person plus 1 cfm/100sf of floor area (assuming a natural leakage average of 2 cfm/100sf) to deal with other house-sourced pollutants, which they believed to require a total of 3 cmf/100 sf to maintain indoor air quality.

            Joe Lstiburek’s and your opinion on whether the ASHRAE standards are excessive does not substitute for the scientific consensus represented in the ASHRAE standard.

          • foursevenfive July 3, 2013 at 2:19 pm #

            It was a consensus – not convinced it was a scientific one though.

  2. Robert Riversong June 18, 2013 at 4:08 pm #

    My comment is truncated. What’s the story here?

  3. 475Floris June 19, 2013 at 4:51 pm #


    You are correct that using OSB or other formaldehyde heavy wood panels on the interior as a vapor retarder can be problematic from an IAQ/IEQ point of view – our solution would be to use INTELLO Plus or DB+ (paper-based) on the interior to avoid these concerns.

    Two have a protected material on both the interior and exterior makes for the best assembly. The vented rain-screen protects both the structure, promotes drying and let’s the siding dry quickly even if it would get wet on all sides. The interior service cavity, protects the air-barrier from the homeowner – allows renovations/rewiring to occur
    without damage to the indeed very tight envelope. The tight envelope assures both proper functioning of the wall, as well minimizes heat loss and maximize control of the indoor air quality (with low flow HRV).

    The ASHRAE 62.2 2013 could indeed overshoot what you actually need for good IAQ and lead to over ventilation, wasting fan energy and conditioned air.

  4. Maria Spinu June 28, 2013 at 10:20 am #

    Hi Ken, interesting blog post. I’m Maria Spinu from DuPont and I’d like to weigh on this discussion.

    You seem to have a strong preference for interior air barriers. Even though an interior air barrier could provide an airtight envelope, there are a number of challenges in achieving continuity with this method. These challenges are mostly due to increased complexity in addressing the high number of details associated with an interior air barrier and are best described by practitioners. I suggest you to take a look at an article published by Walsh Construction Co., a high quality GC who has been delivering airtight buildings consistently, and validated their practices through whole building airtightness testing (Reference: Towards Airtightness – The Contractor’s Role in Designing and Constructing the Air Barrier System, Michael P. Steffen, presented at BEST 3 Conference in April 2012). You can find the presentation at: http://c.ymcdn.com/sites/www.nibs.org/resource/resmgr/BEST/best3_steffen.2.8.pdf

    I would also like to comment on some of your statements:

    “……a common approach we too often see places the air control layer outboard of the insulation layer ……..however, if you only have one airtight layer this is the wrong side – it is a high-performance building turned inside-out”.

    While an air barrier could be located anywhere within the wall (IF it is vapor permeable, so it does not interfere with the drying potential), placing it towards the exterior of the envelope offers the best protection: (1) it protects against forced convection which will bring cold air from the outside into the wall assembly; (2) it protects against convective loops and wind washing, helping retain the insulation R-value; (3) it protects against water intrusion (e.g. from wind driven rain); (4) it protects against contaminants transported by air (either from outside or from degasing of building envelope materials). A second air barrier to the inside of the wall would not hurt, but it would not provide the same level of protection as the exterior air and water barrier.

    You also state that when the air barrier is“…. a continuous exterior shell of OSB or plywood …. this allows conditioned air to readily enter the insulation layer by convective driven air-leaks and diffusion. It also puts a vapor retarder outboard of the insulation and consequently doesn’t prevent warm humid air from reaching cold components, increasing the condensation potential, while the exterior retarder unhelpfully restricts the outward drying potential.”

    Where should I start? Firstly, you attribute control of water vapor transported by air leakage and diffusion to the air barrier layer; the 2 water vapor transport mechanisms are vastly different and require different control strategies. Air transported moisture is controlled with a continuous air barrier; the air infiltration resistance of air barrier materials and the continuity of air barrier systems are critical for effective air leakage control. Vapor diffusion control is achieved with Vapor Barriers, or Vapor Control Layers. These are diffusion closed or vapor impermeable materials. Secondly, the 2 functions, air barrier and vapor barrier are most often decoupled: not only they require different materials properties, but their location within the wall assembly is most often different. Thirdly, an air barrier does NOT need to be a vapor retarder: air barrier materials can be vapor permeable or vapor impermeable. In fact in cold and mixed climates the air barriers installed on the outside MUST be vapor permeable (to allow diffusion drying) while the vapor barrier / vapor retarder MUST be installed on the inside of the wall (in order to control diffusion wetting). For a list of common air barrier materials and their vapor permeance, see http://www.airbarrier.org/

    “A better assembly is possible if we concentrate on how to combine, on the interior, the airtight layer and vapor control”.

    I would say that a better assembly is possible if we decouple the air leakage control from vapor diffusion control: the 2 mechanisms are very different, have different driving forces and require different control strategies. An Air Barrier is REQUIRED in all climates, while Vapor Barriers are only required in certain climates.

    • foursevenfive July 3, 2013 at 9:55 am #

      Hi Maria,

      Very sorry for delay in responding to your blog post, and having your post cut short – I think we’ve finally fixed the bug…fingers crossed.

      Thank you for your comment we look forward to continuing the discussion. Some thoughts in response:

      We agree that an interior air barrier can be more complicated – although if done as John Straube describes in his post it necessarily needn’t be so – so we are as a company spending a great deal of effort to help make it a more easy reality for designers and builders everywhere.

      Thank you for sending the presentation link, it is useful. As Michael Steffen’s notes the IECC 2012 code requirement of 0.4CFM/ft2 of enclosure area. This is a great and important start in our opinion. The reality is that these buildings can and should be much, much tighter – over 10X tighter to meet the Passive House Standard. So as we push for greater and more reliable airtightness the calculation of effort may shift.

      It seems we agree about the benefits of protecting the insulation outboard from the negative effects of wind and weather.

      The air barrier and vapor control may require different strategies or they may not – depends. I think we agree on almost everything else you write about vapor control layers and we understand pretty well various option for air control. We might again suggest if you require vapor control inboard that you take a look at our INTELLO vapor intelligent membrane (that’s also airtight) – of course OSB or plywood sheathing could accomplish this too as we note in the blog post and John Straube seconds.

      Regarding decoupling, we completely agree that everything is climate specific. If you can achieve optimal results with air and vapor control combined, then maybe decoupling isn’t the best option there.


    • lavardera July 18, 2013 at 6:29 pm #

      Maria, If you review the diagrams that accompanied the blog post it becomes quite clear how the proposed wall assemblies overcome the “challenges” associated with achieving continuity. In fact I’d go so far to say that it is easier to achieve continuity in these wall assemblies than it is to achieve on the outside surface.

  5. John Straube July 2, 2013 at 3:24 pm #

    Hi Ken,
    John Straube from University of Waterloo and Building Science Labs here.

    I cant read all of Maria’s blog, nor your response, but I share almost all of her concerns. The approach you take is the several technology generations ago approach, which is why it is still popular in Europe (who generally have been behind Canada and US since the invention of superinsulation). Most leading edge builders in Canada moved on from the service wall on the inside of interior air barrier by the late 80s! This has been tried and found wanting. Too much effort, no real benefits, despite your claims. Exterior-to-the-structure rigid air barriers are the current best practise. This does NOT mean on the exterior of the insulation, which is a classic flaw in understanding. Placing the air control plane on the inside of some or most of the insulation BUT outside of the structure solves any concerns with convection and diffusion, and provides a service space without paying extra for it.
    OSB or plywood or gypsum plus good tapes (like you sell) works as a great air barrier, and adding a housewrap, or liquid membrane (like Dupont sells) turns this layer into a great water control layer. Then add insulating sheathing made of foam, rockwool, or even fiberboard (if you like that risk), means you can have a continous protected air water barrier with zero convection and diffusion risks, with all the benefits of practical continuity that exterior to structure provides.

    • foursevenfive July 3, 2013 at 9:26 am #

      Respectfully disagree on the “who’s in the lead” analysis. I’ve a feeling it’s apples and oranges and feels a bit like saying European cuisine is behind because McDonalds does so well.

      TOTALLY AGREE that the thermal layer defines what is inboard and outboard – and that the air barrier should be inboard of the thermal layer. In seemingly trying to disagree with our post you have described exactly the type of system we are proposing above – including in the detailed wood wall section. (of course we are showing dense-pack cellulose outboard but in our view it could be mineral wool or fiberboard too. No foam, please.)

      I guess we need to be clearer, as we seem in clear agreement as to the basic building science. Of course I’ve a feeling we’ll likely disagree as to what it means to be “cost effective”. Good to discuss.

      • John Straube July 3, 2013 at 10:09 am #

        Of course “whos in the lead” is an opinion. But it is fact that builders and researchers in Canada and the US tried the interior air barrier and service walls in 20 years ago and moved on. That is hardly a McDonalds comparison. I suppose I could call PassivHaus Taco Bell by the same approach.
        Please reread my post. I DID NOT SAY that the air barrier needs be inside the thermal layer I said “some or most”. This is the answer science provides. I did say it needs to be outside the structure to reliably get continuity. This is pratical experience, and the concept you are choosing to ignore. A resource friendly (=cost effective) approach if using wood framing (ie poor conductor) is to fill the stud bay with low cost insulation, and then use rigid exterior insulation over the air barrier. This is not placing the AB “inside the thermal layer”.

        I just read a detailed PhD thesis from KU Leuven (Belgium) which investigated exterior air barriers vs interior air barriers and cellulose filled double stud walls. It confirmed the moisture risk of the “fill the double stud wall with fluffy insulation” because even 0.6ACH@50Pa houses leak sufficient air to cause moisture risks at cold exterior wood elements. Numerous field measurements of exterior insulated air barriers over insulated stud walls have shown that such risks can be essentially eliminated, inward vapor drives managed, exterior wind washing convection controlled, and thermal bridging at framing minimized.

        PS Cost effective to me is the lowest construction cost for the same risk and performance, not comparing high risk walls with low risk at the same price.

        • foursevenfive July 3, 2013 at 11:01 am #

          Really, Taco Bell? Maybe Le Pain Quotidien?

          Agreed “inboard” was an oversimplified characterization. Our details in the post show insulation on both sides of the airbarrier/vapor control layer. Yes “it depends” – with prejudice to the interior.

          Interesting research – will be good to hear more about it.

          What you describe as “using wood framing (ie poor conductor) is to fill the stud bay with low cost insulation, and then use rigid exterior insulation over the air barrier.” falls completely within our description in the post – although we’d prefer to see mineral wool or fiberboard.
          We don’t show that detail explicitly but we’ll gladly include it.

        • lavardera July 18, 2013 at 6:04 pm #

          John says:
          “I DID NOT SAY that the air barrier needs be inside the thermal layer I said “some or most”. This is the answer science provides. I did say it needs to be outside the structure to reliably get continuity. This is pratical experience, and the concept you are choosing to ignore.”

          Continuity of the air barrier is easily achieved in the Swedish housing industry where it is installed inside the stud wall structure. Its done simply, and quickly, and largely without tape. This operation is much faster and less expensive than the exterior side insulation assemblies you have described.

          There – now you don’t have to ignore that fact, nor be ignorant of it any longer.

          • J Jones July 6, 2015 at 8:52 am #


            Can you provide a link or more information about the Swedish approach to interior air barrier placement? How are they securing the AB within the stud wall structure? Is this outside of the stud wall insulation and inside of the exterior sheathing?

    • lavardera July 18, 2013 at 5:47 pm #

      I have to weigh in with 475 on this. John is probably one of the best Building Scientists in North America, yet this comment comes across as dismissing decades of building science work in the EU.

      Any casual observer of the “green” building movement can observe a flow of innovations from the EU to North America, whether its the blower door, or high performance windows, energy recovery equipment, or sophisticated membranes. In fact I’ve never heard anyone, nor had a discussion with anybody that considered North America “Leading” as you just characterized in that comment. I can not over emphasize how astonished I was when I read that..? Wow.

      Multi layered wall assemblies in the EU similar to what is described in this blog post are the result of decades of rigorous building science study, fueled by feedback from an industry that has actually implemented and built millions of square feet of such “superinsulated” walls. By comparison the characterization of the wall system you describe as “leading” seems bizarre, if not self-gratuitous. This is “leading”? An industry that can’t get out of the way of Energy Star?

      The basis for your claim is that it costs more to build a service cavity? I’ll propose that the creation of a service cavity with the application of a simple furring layer to create a service cavity represents much less construction cost than the taping of sheathing you describe, and the addition of several layers of foam boards to achieve the “most” of the insulation condition you describe, all done on ladders on the exterior of a wall, not to mention the complication and additional effort these exterior layers create at the window openings, all of which again must be resolved on a ladder. It seems painfully obvious that the service cavity inside the structure is easier, faster, simpler, yet you characterize it as “too much effort” and “no real benefit”. Has this been quantified that with a study? Lacking data I have to say your claim sounds incredible.

Leave a Reply