Need an opinion on the practical side of perm ratings
MarkM3
| Posted in Energy Efficiency and Durability on
Thanks so much for your time and advice. I plan to build in a high humidity environment, with off-grid PV power. Central dehumidification is a must, and is by far the biggest electrical load. I don’t have a feel for whether plywood sheathing/OSB (perm range of 1-4?), combined with large wall surface areas, will allow a significant amount of moisture to creep into the house 24/7, requiring significant additional dehumidification beyond what we’d expect from normal interior sources (humans, showers, cooking, open windows, etc). Or, put another way, at what perm level can I maybe discount the exterior moisture transfer as rounding error?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Air leakage and ventilation air is the big moisture-mover into the house when living in a tropical swamp, not vapor diffusion through the sheathing.
ANY vapor diffusion would be a rounding error compared to the moisture introduced by infiltration + ventilation.
For the record, plywood & OSB have variable vapor permeance, that changes with the RH of the proximate ar, and the moisture content of the material. At very high RH plywood's vapor permeance rises faster than OSB's but for your purposes it's a "who cares?" proposition.
https://www.energyvanguard.com/sites/default/files/hubimage/bsi-038-mind-the-gap-permeance-of-plywood-and-osb.jpg
For purposes of keeping indoor air humidity bounded it's really ALL about the air transported moisture. Build it tight and ventilate only as much as necessary.
May I ask where this off grid building will be located?
Dehumidifier can draw a lot of current. What is your power source?
Yes, absolutely. I have an entire tome on the subject. I'll just paste it here so you can get it all on one dose. I am starving for good advice. Very much appreciate any contributions.
In light of your comments, I might lean toward the Huber Z-zip system with some insulation. It's not a barrier, and we don't care. Would fix the condensation risk, and I could have higher humidity.
The build environment at 4000' elevation in the Kailua-Kona Hawaii "cloud forest" is a very unique environment. First off, a cloud forest is different from a rain forest in that a cloud forest gets a substantial amount of the water from condensation. The forest is usually running between 70% (afternoon) and 100% (overnight) RH.
We would actively dehumidfy the house via a central dehumidification system, with ducting similar to a mainland house, except the source ducts for the dry air would inject into wet areas/closets/etc as opposed to near windows.
From a temperature point-of-view, this is one of the most benign environments in existence. A typical high of 75F, typical low of 55F. Add in a bit of solar heat gain (lots of windows), and I think that we're right where we want to be. There is very little winter/summer seasonal variation.
So, we are not highly motivated to add excess thermal insulation. With the planned 2x6 construction, a nominal R19 should be more than adequate, and our overall thermal performance will, of course, be heavily driven by large window areas of R4 and thermal bridging.
On exceptional days, I understand we can get frost. I would like to design a wall system that performs down to ~35F.
One important point - the house is full off-grid solar. So while we might reflexively shoot for an interior relative humidity of 50-60% to avoid mold/mildew, in colder weather this humidity level would condense on the interior of our wall sheathing. It would solve all my concerns if I knew that I'd have enough electric to power where I could bring the interior RH down to 30-35%.
I am intrigued by the idea of humidi-mass (is there a better term?) - analagous to thermal mass. If the entire wall structure of drywall, insulation, studs, and sheathing can store and buffer moisture, then perhaps I can over-dehumidfy on strong solar days, and have some intertia where the wall will accumulate moisture rather slowly on stormy days. The solar model says that on roughly half the days, I'll have more energy than I know what to do with.
The reference bogey for wall construction, from interior to exterior would be:
Latex paint (no special vapor retarder)
5/8" type X sheetrock (possibly standard 1/2")
Dense pack cellulose (possibly fiberglass, unfaced)
1/2" plywood (possibly Huber Zip system)
Tyvek
Keene rain screen (possibly 1/4" furring strips)
Boral Tru Exterior siding (no capacity for moisture storing, and rated for ground contact)
The weakness of this scheme is that humidity from the forest continues 24/7 to diffuse through the wall, potentially requiring a significant electrical draw to repeatedly dehumidfy. (I think you just provided a good counterpoint on this - which may somewhat invalidate my further comments)
Alternatives considered:
- Add extra foam insulation outboard of sheathing, behind the Tyvek, allowing the condensation surface on the inside of the sheathing to be at a higher temperature. This is the current best practice in very cold climates, but it is expensive, challenging to flash and properly manage drainage, and local build teams are not so familiar with this scheme.
- Closed cell foam insulation. Technically ideal, but this is stupidly expensive in this neighborhood.
- Modify the bogey wall above to coat the exterior of the plywood with a vapor barrier (not retarder). Seal the entire envelope tightly against both air and vapor. Since the condensation risk exists even with uncoated plywood, we'd still need to either insulate outside of this barrier, or dehumidify to lower interior RH levels.
- Install a layer of plastic sheeting vapor barrier behind the drywall, and let the wall dry to the exterior. "Dry" is a relative term... the entire wall structure would gravitate to the forest RH average - 85%? - too wet for me. Also, the perm rating for plywood is not that high, so drying would be very slow.
- Put a vapor retarder (not barrier) coating on both the interior drywall (via paint) and on the outside of the sheathing. But the cavity in between would still gravitate to average RH levels, and be very slow to move from that. Doesn't help at all with condensation risk.
- Two-layer sandwich of fiberglass insulation, with the outboard layer having a Kraft face on the interior side (insulation/Kraft/insulation). Complicated, and not so easy with standard sizes, as two 3.5" R11 batts substantially exceed the 5.5" stud cavity with 2x6 framing. Maybe if it was cheap enough to frame with 2x8s? But the Kraft facing is still a retarder, not a barrier, and the outboard insulation as well as sheathing would gravitate to the forest RH average.
- If allowed by code in this seismic environment, maybe a Huber Zip System - say, a 1.5" thick R6 - for the sheathing instead of plywood? Should be enough "R" value in that skin to keep the interior surface warm enough to avoid condensation. However, since the coating is still a retarder, we would need to apply a vapor barrier coating to - well, I'm not clear that this is even an option. I have no feel for whether 1 "perm" is going to lead to a significant amount of moisture infiltration, but I assume for now that any perm >0.1 would interact with the huge surface area to continue to bring humidity into the house.
In conclusion, after exploring these alternatives, I propose that we build the wall as outlined above, but with the addition of a vapor barrier (something like Prosoco R-guard VB) coating on the outside of the skin, and behind the Tyvek.
This barrier would have to cover the entire house, and would have to be well-executed. In our favor, the result is highly visible for inspection.
This means that we'd need to keep the house interior RH closer to 30-40%, rather than in the 50-60% range. However, if we are not weeping moisture throught the walls, and are vulnerable only to normal human perpiration, showers, open toilet bowls, laundry, and the occasional open window - perhaps this is really something we can accomplish. We do have a wood burning stove to assist on colder days.
Here's another even simpler way of thinking of things... whether the wall is permeable or not, vapor will condense. Doesn't matter if there is a coating outside or not. This boils down to two variables... keep the condensation layer warm, by moving it inboard - or more practically, insulating outboard of that. Or drop the humidity level inside.
Thanks much, Jon. Yeah, I've kicked around "kitty litter on cookie sheets" as another way of buffering. Kind labor/space intensive, but if you knew a big storm was coming... maybe. I plan a two stage build of a 12x24 2-story shed first - shop below, "Hillbilly Hilton" above. I'll live there and get data on what works, and apply that to the house. Mainly, just don't want to implement something that fits into the long list of "seemed like a good idea at the time" schemes.
Agreed, it's not clear to me how one creates a practical large exposed surface area of bentonite and just how well it works.
In a cloud forest it's probably better to not use moisture susceptible wood framing or wood sheathing, and build with moisture tolerant materials such as stone, masonry or concrete. A SCIP approach can work well in Hawaii, with a 2-3" EPS core and wire-reinforced sprayed concrete (or hand applied concrete, as is often done in remote areas) on both sides. A thin concrete monocoque like that would also be easier to air seal, and even 2" of EPS would be under 2 perms (but it's still the infiltration that matters.) Unlike wood, concrete performs just fine when saturated, and the EPS would not become saturated if the interior is being dehumidifed even part of the time. EPS works just fine as dock floats, even if below the water line there is enough water in the intersitial spaces to cut into thermal performance. (Even fairly saturated EPS delivers R2-ish/inch performance. In US climate zone 1 with concrete wall a continuous R3 or R4 meets IRC code minimum.)
The Hi`ilani EcoHouse is an off-grid SCIP construction building (but not in the cloud forest) on the big island. Some pictures that house as it was being built live here:
https://inhabitat.com/studio-rmas-hiilani-ecohouse-is-americas-first-carbon-neutral-concrete-structure-and-its-volcano-adjacent/
I don't have direct experience, but would imagine that most of the structural wood will become nearly saturated much of the time in that environment, and even treated would could become compromised over time.
Last time I did the calculation, I came up with 5 perms as my "don't worry about it level" of moisture diffusion - focus mainly on air infiltration and ERV. You can store dehumidification (which makes sense with solar PV) with materials like bentonite. But a good drying every other day will prevent mold growth.
Mark,
How are houses typically built there? I would imagine modifying a construction type that has a pretty good track record over time might be safer than introducing a whole new way of building.
Dana, thanks so much for continuing the conversation. I suspect the Hiilani House is in a very similar environment, but I strongly suspect that we are in different leagues regarding budget. It's hard to see this SCIP approach, complete with "fly in crews" as being anywhere near cost competitive with pressure treated plywood or the Zip-system. Stone, concrete, masonry are rarely used here in residential construction, due to cost. I suspect the sheathing would indeed gravitate right to the forest average humidity of around 85%. I think the challenge now is the most cost effective way of getting some R-value outside of the first condensing surface. I don't know if the Zip system makes sense - whether that comes in a pressure treated version, combined with insulation, or if I'm just going to have to bite the bullet and put some TBD insulation outboard of normal PT plywood. Thanks again, Mark
Malcolm, thanks much for your thoughts. Yes this would be the first instinct, but in this region we move from micro-climate to nano-climate. The vast majority of houses are at lower elevation where it is substantially dryer and hotter. So there are very few relevant examples to fall back on. At least I haven't found any that have had any sort of long-term impartial test/evaluation. Closed cell foam seems like a credible option, if one had a few extra bundles of $10k stuffed in the mattress. Or coating the sheathing with a true vapor barrier, with an insulation layer outboard of that. Joseph Lstiburek's "the perfect wall" seems like an excellent fit - I guess I'm just having a hard time with the cost and complexity of install (and local knowledge of best practices) with the outboard insulation layer.
Mark,
That's just my boiler-plate advice for anyone - to see what is already working in their climate. But I live in an area where conditions can vary enormously within a few miles too, so I appreciate you may have to look for a different solution. I have no experience at all with building in the tropics. Hopefully someone here will be able to offer better suggestions. Good luck with your build!
Mark,
It looks like you might have access to autoclaved aerated concrete. It’s a diy friendly product and probably would address the concerns Dana raised in his post.
Thanks, Steve - I'll dig into that.