Roof sheathing in the PNW
Here in the damp PNW roof sheathing on well ventilated and air-sealed free-framed and trussed roofs often exhibits some discolouration especially on the north side. It is seldom serious, but it does indicate the sheathing is damp enough to sustain a small amount of mold growth for a large part of the year.
I remember this problem came up in a discussion with Kota Ueno of BSC at some point. I wonder if their research lead to any interesting solutions?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
I don't remember the GBA thread you're talking about. But in one of my articles, All About Attic Venting, I including a photo (Image #2) showing the phenomenon you are describing.
The photo came from a paper called "Highly Insulated, Ventilated, Wood-Framed Attics in Cool Marine Climates" by Patrick Roppel, Neil Norris, and Mark Lawton.
Here is a link to the paper: "Highly Insulated, Ventilated, Wood-Framed Attics in Cool Marine Climates".
I found two more links to the research they were doing in Vancouver.
Perhaps some non-permeable film (eg, foil) adhered to the underside of the decking would prevent such condensation from entering the wood.
Would also be interesting to explore a sealed and dehumidified attic where the attic floor remains the thermal barrier.
Active humidity control in unconditioned spaces is hard, because temperature swings without air flow can lead to relative humidity swings (at a constant humidity ratio). But in some climates, with good controls, and good air sealing, it might work. Marine climates don't have huge temperature swings, and so they are good candidates.
I've also seen a couple of systems that vent such spaces based on dew point sensors inside and outside, in order to bring in outside air only when it will be beneficial for lowering absolute humidity. I am kind of intrigued by that, but any active solution is likely to fail at some point, and is probably most likely to fail after the energy nerd has moved out. The new residents are unlikely to notice that there is anything wrong until the rot gets worse than it would have been with conventional passive ventilation. So such a system is not likely to be a good idea unless you are mitigating a serious problem.
One solution that worked well when the roof was clad in metal was using strapping rather than sheathing, but the new seismic code requirements preclude that now.
As the Vancouver test huts show, in the PNW the conditions for mold growth exist in well ventilated roofs independent of insulation or air-sealing problems. I have some minor mold growth on both my shed roof, and the wood siding on the north face of my house. The extent seems to be dependant both on solar orientation and the presence of trees and vegetation near the dwelling. The houses I have built on sunny, cleared lots don't show any variation between the north and south sides. The difference is more pronounced in those on shadier, more treed lots.
I agree. We have enough building envelope failures in our existing housing stock due to the lack of even rudimentary maintenance. The solution needs to be a more robust and forgiving roof system, not one dependant of active controls.