I just heard rumour that Canada is looking at some building code changes that would include insulating the roofline and conditioning the entire building envelope as opposed to insulating the attic floor and venting the attic. We are mid way through a major renovation of our 110 year old home and planning to use a closed cell foam to seal the attic floor. Should I consider using the same foam to insulate the roof line as opposed to the attic floor? What is the reasoning behind this approach and is it a better option?

the reasoning is:
the idea that we can have an accessible part of the house, like the basement and the attic, where the access doors are uninsulated, unweatherstripped, and the walls & floors between have numerous penatrations and are typicallly not well sealed, and in the case of most stairways, uninsulated,
is a fiction. Passive House brought to this continent the concept that the whole envelope needs to be well sealed and insulated from the outside air. If you want a "cold" attic, you can do it, but treat the whole space as well as the access points and stairs the same as you would exterior space, with thermally broken framing, good air sealing everywhere, and the same levels of insulation. (so you would need R50-60 under the stairs, for example) Fairly straightforward to eliminate "storage attics" in new homes, but harder to do in existing homes, so in most retrofit cases the roof is insulated and brought into the thermal envelope. While you are in fact adding cubic footage to your envelope, the added heat requirement is usually less than the heat loss from an uninsulated attic.

Posted By Bob I on 14 Apr 2014 12:29 PM
the reasoning is:
the idea that we can have an accessible part of the house, like the basement and the attic, where the access doors are uninsulated, unweatherstripped, and the walls & floors between have numerous penatrations and are typicallly not well sealed, and in the case of most stairways, uninsulated,
is a fiction. Passive House brought to this continent the concept that the whole envelope needs to be well sealed and insulated from the outside air. If you want a "cold" attic, you can do it, but treat the whole space as well as the access points and stairs the same as you would exterior space, with thermally broken framing, good air sealing everywhere, and the same levels of insulation. (so you would need R50-60 under the stairs, for example) Fairly straightforward to eliminate "storage attics" in new homes, but harder to do in existing homes, so in most retrofit cases the roof is insulated and brought into the thermal envelope. While you are in fact adding cubic footage to your envelope, the added heat requirement is usually less than the heat loss from an uninsulated attic.

And here I was under the impression that the whole PassivHaus approach was derivative of a home grown Canuckistani design in Saskatchewan that predated the first Darmstadt PassivHaus by more than a decade.  SFAIK Regina, Saskatchewan has always been located on "this continent".     (What goes around, comes around?)

You're right that many of the "early" North American superinsulated houses were used as a basis for Passive House, and that that one in particular is often cited. I don't know what each of them brought to the discussion, but it seems pretty clear that what PH did was to identify the contribution of particular features of these houses and combine them to generate metrics that work and are repeatable.

Posted By Bob I on 14 Apr 2014 05:47 PM
You're right that many of the "early" North American superinsulated houses were used as a basis for Passive House, and that that one in particular is often cited. I don't know what each of them brought to the discussion, but it seems pretty clear that what PH did was to identify the contribution of particular features of these houses and combine them to generate metrics that work and are repeatable.

Even the Native American Indians residing in the desert SW understood thermal mass and utilizing stack effect for cooling. It was crude and not without issues but they did use techniques that can be traced back to that building style.

Zad - have not heard anything about that as yet. On some of the houses, it would add 10,000 to 20,000 cubic feet of conditioned space as well as increase the exterior building envelope by 10 to 40 percent. That is a lot of wasted energy for heating the attic space. I would be very surprised if the Canadian codes went that way.

Course anything that keeps the cost of nat gas up is a good thing, right? ;-)

Our house is hipped so the added conditioned space wouldn't be that great. In fact, it would not even add that much to insulate given that the roofline is less distance than the attic floor and exterior wall in the attic. Although, I would have to go much deeper with the foam than I would if I did a seal coat and blow in for the attic floor. Hmm, interesting option and one that may be a better choice for our home. Is there a risk to shortening the life span of our asphalt shingles?

The PassivHaus Institut was able to better measure & model the energy use due to the better modeling tools and more research data on building materials available in 1990 than in 1976. A circa 1990 desktop PC had more data handling capacity & compute power than many mid-1970s mainframes.

But the air-tight & well-insulated concept wasn't invented in Darmstadt, only pared down and refined, and given specific targets, and popularized as a standard. Air tight and well insulated was very much included definition of designer Harold Orr's goal a-priori, and he had to argue for it:

"My prescription was at least R40 walls, R60 attic, triple glazed windows or windows with shutters, no basement, a crawl space with R20 in the floor system, and a very tight air / vapour barrier."

"If we could reduce air leakage by 80% and heat loss to ground (basement) by about 80% we would have a 64% reduction in heat loss without touching the windows and doors, walls, and ceiling. If we use 6 times as much insulation in the walls and ceiling and use much better windows and doors, we would be down to a total heat loss that is about 20% of the heat loss of a conventional house."

http://www.ecohome.net/news/latest/principal-designer-house-inspired-global-passivhaus-movement-reflects-project-started

By most measures the Saskatchewan Conservation House was over-built to ensure success in the face of large known & unknown error bars. The original active solar thermal space heating was later removed since it was never actually needed, but they weren't so sure at the time. They also weren't too sure about how efficient their experimental prototype HRV system would be over time, and how much ventilation was actually needed.

I think it's fair to say that the approach was invented in Canada, even if it was the Darmstadt crew who developed cheap & easy design tools for reaching (somewhat arbitrary) target energy use levels using that approach, selling it as a standard.

Posted By Zad on 14 Apr 2014 11:38 PM
Our house is hipped so the added conditioned space wouldn't be that great. In fact, it would not even add that much to insulate given that the roofline is less distance than the attic floor and exterior wall in the attic. Although, I would have to go much deeper with the foam than I would if I did a seal coat and blow in for the attic floor. Hmm, interesting option and one that may be a better choice for our home. Is there a risk to shortening the life span of our asphalt shingles?

The color of the shingles has a larger effect on peak/average shingle temperatures and shingle life than vented vs. unvented.   The shingles shed heat primarily by re-radiation back at the sky and convecting exterior air, not from heating up the attic air through ~R1 of underlayment. That roof deck ventilation factor is measurable in the shingle temperatures in careful side-by-side experimental comparisons, but it's a second-order effect of low to no consequence.  A change in latitude of 10 degrees makes a far bigger difference.

A primary benefit of going unvented is how reliably & completely one can air seal the top of the building, reducing stack effect infiltration to near zero.  With a vented attic you have to air seal at the attic floor plane, which typically has an order of magnitude more penetrations. Also,  things like wind currents and the ratio of ridge/soffit cross sectional area can depressurize the attic relative to the conditioned space below, further driving infiltration.

Ok then, we are going to insulate the roof...seems like an efficient way to go and it will allow us to have a ton of heated storage. Very cool. Thanks for all the input.

I specified a hot roof in my own 1921 renovation and several houses since. No tax on the shingles, as Dana argues, and no penalty for heat load, also pointed out in Dana's reference to infiltration and further illuminated by Bob. Not only air tight from the shingles down, but no thermal bridging, increased strength and total access to the attic. Though I won't store anything in attic, I can change my mind about wiring, ventilation, etc. Bravo

Whoa Nellie, not so fast. At insulation inspection I spoke with the building inspector and she will not pass a conditioned attic. Apparently the code leaves enough room for ambiguity that if the inspector is not savvy or comfortable with it they can fail a non-vented attic. She was adamant it was not something she would pass.