Venting an interior service cavity
TLDR Version:
Build thick well-insulated wall. Use interior air barrier / vapor retarder.
Add interior service cavity. Make it double-batten.
Vent interior service cavity to inside air through holes in floor & ceiling trim. Convective loop behind drywall ensures steady airflow and ensures that air barrier is close to moisture+thermal equilibrium with interior air, which permits drying to the inside and hygric buffering of interior humidity.
Now you don’t have to worry what inhabitants do to paint/cover drywall, only whether they leave the trim intact.
Point The First:
So you have your super-insulated wooden house that can dry to the inside or the outside or emit moisture to the special level of hell they reserve for child molesters and people who talk at the theater. It’s made this way because insulation slows down drying, and if any water gets inside and can’t get out, your walls rot and your house falls down. Number one best way to destroy your air conditioned and heated house is apparently any kind of vapor barrier, or anything that retards vapor too much. That means rummaging through your GC’s van for their supply of 6mil poly and lighting it on fire at the edge of the lot with a ‘No! Bad dog!’ expression. It means you can’t have vinyl wallpaper. It means only certain types of posters and wall art. It means you can’t have anything plastic flush up against the wall. It means you can’t have oil paint. It means you better be careful where you put foam (particularly closed-cell), it means you better be careful how and what sort of OSB you use, it means watching an hour-long presentation by a man with an unpronounceable last name and an indefatigable sense of dad-joke humor on the topic of “Moisture Management”.
Point The Second:
So you have your super-insulated wall made of cellulose, rockwool, and engineering daydreams. In order to avoid puncturing the expensive imported artisan molecular interior air barrier (which has to be laid over the Insulweb holding back the DPC) with drywall screws, picture-hanging nails, whole electrical boxes, wiring runs, plumbing, et cetera, you want a service cavity.
Since we will need to puncture it in order to install anything over it, we will want to seal those puncture holes as best as we can. How do people seal things at an interface? What do submarine hatches use? What’s holding the vacuum out of the ISS portholes? Why does my garden hose work? Generally it seems to come down to something rigid compressing the hell out of a sheet of rubber braced against something else rigid. Looks like we have butyl caulking tape / sealant tape already being used in the industry for other things. Once compressed, it will create a gasket around fasteners. Just need to have some screws in there for pullout resistance. Does 475 sell it (indicating that it makes sense in a European Passivhaus market)? Yes! Pro Clima TESCON NAIDECK. I suppose you could use something fluid applied, if you wanted to get messy, but let’s not.
Should it be vertical or horizontal? Vertical gives you easy connection for electrical but makes wiring runs difficult, while horizontal will give you undesireable Chicago Outlets ( http://www.city-data.com/forum/chicago/2803166-horizontal-outlets.html ). Why not both? Horizontal in the back, vertical in the front to attach boxes to without any blocking (and permit vertical wire runs, which would otherwise require your electrician to own a drill). Use 2×3’s in back and 2×3’s in front, both laid flat, for 3″ depth (or do the 2×3’s in front on edge for 4″ depth).
What can we do with that space? What else does this let us do?
Hrrm… Let’s revisit Point The First. I may be able to avoid vapor barriers… in a room? But it’s going to be tough to police kids and future homeowners with perfect reliability. I can’t ever be confident that a tenant isn’t going to proclaim their love for the Golden Age of Serial Television ( https://www.facebook.com/frenchguycooking/photos/le-studiojust-got-better-screengrab-from-my-latest-upload-super-excited-about-20/1547904195328253/ ) by destroying the wall I built. I don’t even want to think about storage rooms – does everything there have room to breath?The consequence of screwing it up is invisible right up until “Major mold remediation”. I can’t ever be confident that a fix-and-patch job is going to preserve the integrity of the exterior building envelope. Furthermore, I don’t *want* to have to figure out the perm rating for a nephew’s fingerpaint or a tile wall or Arbitrary Fashionable 2028 Trend In Wallboard. I’m try’na build this thing for a century lifespan, and it can’t presume that I’m there blowing the lifeguard whistle that whole time. When wall damage was done in my childhood home, the contrasting paint was on there layer after layer after layer. I don’t know how much was latex versus oil, or what that thick strata of primer and finish coats would rate on a perm scale.
So we have all this airspace behind the drywall. Could we use… that… space… to dry the wall? How does drying work, anyway? Molecular diffusion, which is extremely slow unless large gradients exist, and air movement, which is extremely inhibited by insulation. Once the moisture gets inside the service cavity through the vapor-retarding air barrier, it has to leave through the… um…. It has to build up a concentration on the drywall and percolate through it. Which probably means that the service cavity will have high humidity during a drying event (when an Act Of God storm meets an American Windowframe, say). How do you fix that?
Also, part of the purpose of all this cellulose was to permit additional hygric buffering capacity – if the bathroom vent fan fails and I’m too much of a deadbeat to replace it (or we have a roof leak, or Little Timmy leaves the screen door open in a thunderstorm, or a houseguest tries to mop a hardwood floor…), the walls should have a little space to hold water (to suck water out of the air) without maxing out their capability to hold water and rotting. How is this hygric buffering supposed to be exploited if they’re being sealed behind not only a vapor retarder, but also drywall/plaster/paint?
Convection is a ventilation engine that runs 24/7 totally reliably as long as heat can transfer from hot to cold, . It will move air faster the taller the venting stack is. What if I just ran that convection through the front of the service cavity? Drill out holes every inch in the baseboard and crown molding (or tell a finish carpenter to vent it and figure out how to make it look pretty), and you create a convective loop inside the wall that slowly transmits moisture regardless of interior decorating. It’s well-segregated by the air barrier (not to mention the dense pack cellulose, which is a mediocre but redundant air barrier itself) from the liquid dew point and/or the frost phase transition as you get deeper(colder) in the wall. If you cover up the wall, air moves in horizontally along the service cavity to convect away any differences. You are not afflicted with problems like “The pile of blankets and mattress foam you’ve got in the corner of your bedroom functioned as such effective insulation that your carpet froze solid, and all winter of it accumulating moisture, being run by vapor drive, created a fungal colony in the spring whose mycelial network touches all things in the multiverse”.
It’s a bit like a vented rainscreen or a vented roof cavity, but it keeps the drywall and the insulation at about the same temperature and moisture content for the same interior and exterior conditions, regardless of what the homeowner does to the interior.
All the advice I see is to avoid convective loops in the wall, but presuming that the walls & insulation are Really Overbuilt(tm), and outboard of a service cavity, can you imagine it hurting anything? Or helping?
PS: I understand service cavities will Eat Up Interior Square Footage. All good things do
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I'll admit that my attention span isn't probably what it should be - in part due to the internet - but any chance of a Cliff Notes version.
Sorry, brevity requires some conscious effort on my part:
TLDR Version:
Build thick well-insulated wall. Use interior air barrier / vapor retarder.
Add interior service cavity. Make it double-batten.
Vent interior service cavity to inside air through holes in floor & ceiling trim. Convective loop behind drywall ensures steady airflow and ensures that air barrier is close to moisture+thermal equilibrium with interior air, which permits drying to the inside and hygric buffering of interior humidity.
Now you don't have to worry what inhabitants do to paint/cover drywall, only whether they leave the trim intact.
If you are going to go all out on wall moisture, you need to pressurize the wall in Summer and depressurize it in Winter - air movement (through the wall) is far more important than vapor diffusion - for both wetting and drying. So install a few reversible fans between the interior and a sealed service cavity. But what about Spring and Fall? Install a up-high diffusion port for that - clearly marked "do not paint".
I have honestly thought about an active assembly.
Assuming active water somehow gets in to a wall built with the dreaded double vapor barrier assembly, how do you get it out?
I think the way I would do it is a double vapor barrier on the outside, and an empty space (or a space filled with very low-density insulation) inside with pinholed 2" flex duct on the bottom to supply dry air, and pinholed 2" flex duct on top to collect it. Run this through a dehumidifier and fan in its own thermal envelope.
Less fancifully: If for some reason fossil fuel prices go through the roof, we are going to be looking at significantly higher amounts of wall insulation, and we are going to be taking moisture seriously. What works at R-10 doesn't work at R-20. What works at R-20 doesn't work at R-40. What works at R-40 doesn't work at R-80.
How will you clean the ventilated service cavity?
No idea. Would it definitely be necessary? Anything I can imagine to filter this system would require overly much maintenance (people assure me that if anything in a home requires regular maintenance without obvious indication of immediate costly damage, you can count on it being neglected). Cleaning seems tough.
Maybe if instead of vented trim you used an un-caulked vertical shiplap wainscoting over the service cavity extending up from the floor to a chair rail? That would probably leak enough air to do it if it never got painted over, and if it was friction-fit into stiles at the baseboard and chair rail you could remove it to do work on that bottom 2-3' of the wall without destroying drywall, maybe even tool-lessly.
Tim,
Don't worry about cleaning it. The mice running through the wall will clean it for you :-0
Hi Tim -
If you add this interior "chase" and don't seal it, there will enough penetrations to couple it to interior conditioned space. I think you are over-thinking this space SO LONG as to the exterior of this chase you have a high performance assembly: continuous water, air, and thermal control layers.
Peter
Peter Yost,
The idea is to have this air space sealed on the wall side with INTELLO PLUS (or comparable). It would be _potentially_ sealed on the interior conditioned space unreliably - with airtight drywall, with posters or wallpaper, with oil-based paint, with whatever. We can't predict that at build time, and we can't enforce it afterwards. The idea of vented trim is to make it easier to couple it tightly to the interior conditioned space, even with an uncooperative inhabitant, and maintain the ability to rapidly dry to the interior through a large surface area of INTELLO PLUS, without permitting significant air leakage through the wall.
The service cavity needs to be there anyway in order to airseal tightly and still run trades. It needs to be thick in order to permit normal electrical boxes and plumbing fixtures.
It's not so much about whether this is strictly necessary, it's about resilience and endurance. It's not hard to build a wood house that lasts a hundred years. It's not hard to build a wood house that's super-efficient. It's quite hard to build a wood house that's super-efficient and lasts a hundred years. It's extraordinarily difficult to build a wood house that's super-efficient and lasts a hundred years even assuming homeowner mistreatment, mistakes, or renovation... But that's the bar. Some redundancy in things like air barriers should certainly help.
The design I am pursuing (on paper/CAD) is a double-stud wall relying on mostly cellulose for quasi-Passivhaus performance. I keep tossing around design ideas like this to refine it for performance and buildability, but presently it looks like drywall/plasterboard, vertical 2x3's with airgap, horizontal 2x3's with airgap, intello plus, insulweb, 5.5" cellulose with vertical 2x6's, half inch taped ply, another layer of half inch taped ply, 5.5" cellulose with vertical 2x6's, insulweb, horizontal 2x3's with rockwool, WRB, vertical 2x3's with airgap, Hardie Plank. Overall R-value of about 44 neglecting thermal bridging. Optionally I could add more rockwool to the parts of the service cavity with no utilities run (which is the null hypothesis if venting doesn't make any sense). Avoiding thermal bridging and feeling like the house is fairly well-braced against wind/seismic and extremely well-protected against gradual moisture damage were drivers to this design, but I expect I'll refine it more as I learn more.
Peter Engle,
Rodents are indeed a challenge here with service cavities. They're a challenge with cellulose in general. They're a challenge with rockwool, fiberglass, and foam in general.
They can not only chew through things to find food, they are literally named for the latin 'to chew' - if they stop chewing through things their teeth overgrow and they die. I don't understand how you build a high-performance home that's rat-proof, even assuming no air gaps. I don't think that venting trim makes walls any more susceptible, but it doesn't help either. Any ideas?