Vapor transmission in wall assembly
I’m trying to puzzle out a wall assembly for a cabin that I’m building, and struggling with a vapor transmission issue that I’d like some input on. First time poster – long time reader
Some info: I’m in the northeast (zone 6). I intend this to be a weekender, and being the conservationist and cheapskate that I am, I do not intend to heat it continuously. Best view and exposure are to the south, and though I have a pretty good tree canopy overhead, I anticipate that I’ll get reasonable winter solar exposure. I’m designing quite a bit of thermal mass (slab on grade and plastered straw bale), and quite a bit of south-facing glazing, so I’m hoping that I’ll stay above freezing through the unheated spells (though I’m sure that I will still freeze protect).
The bulk of the wall system is straw bale with lime plaster inside and out. I’m fairly comfortable with this in a go cold scenario. The lime plaster is vapor permeable, the straw has fairly robust vapor storage capacities (that’s vapor storage, not water), and I can dry to the inside or the out. The straw bale sits on a toe-up/knee wall: 6” for part of the house (about 40% of the perimeter), 24” for the rest (about 60% of the perimeter. This 24” knee wall is my struggle. Outside-in, that knee wall assembly is approximately 10” of rubble masonry, an airspace (optional), an insulated stud wall, lath and lime plaster. The knee wall sits on a concrete bond beam, fyi.
Given that this is rubble masonry, I cannot insulate from the outside – that would defeat the point of using the natural stone. My slab insulation marries to the insulated stud wall and straw bale sill for insulation continuity. Also given that this is rubble, I want to ensure that I’m protecting the stud wall assembly from air and water (water, not vapor) intrusion from the outside. Lastly, I want to make sure that I’m allowing the natural vapor drive (inside to outside) to occur so that I’m not forcing drying only to the inside. I think I’m to the rub.
My first pass at this was (inside-out) lime plaster on lath, on studs insulated with cellulose, 1-2″ rigid xps, airspace for weeps, rubble masonry. But I’m concerned that I’ve built a condensation surface at the xps. As it goes cold, the inside of the xps will certainly fall below the depoint, so I’ll have moisture (frost) building up in the insulation cavity. I like the cellulose for its vapor storage capacity, but I can’t image that what I’m describing is a good thing. Problem is, if I omit the xps, I solve the vapor problem, but I’m allowing thermal bridging, and more importantly, I’m not protecting the stud wall or cellulose from air or water.
I have considered spf in the stud cavity – if closed cell, I could keep the xps on the outside and the airspace, if open cell, I would not want the xps (lower permeability as you move outwards), so I suppose I would spray against the back side of the rubble wall. That seems somewhat questionable, as I do think I want to weep the back surface of that rubble wall. All that said, I’m very hesitant about using spray foam based on my own sketicism about off-gassing and things that I’ve read here and elsewhere.
Thoughts? Any help from your experience would be greatly apprciated. Always best to do it right the first time!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Do you have a drawing of this assembly, Drat? It may be helpful.
If straw bale, then why anything else? Maybe I am not getting the picture........
picture (or drawing) is worth a thousand words, so let's see if i can figure out how to attach...
Looking at the detail above, you may wonder why I'm doing this. As a standalone detail, it doesn't make much sense. So some more background - the house site is sloping at about 1:10. I enter on the uphill side (north), and the major living spaces are on the downhill side (south). I want higher ceiling heights in the living spaces, and I want to encourage access to the outdoors from that side, so the slab steps 18" down. On the north side, I have a simple (ish) 6" toe up, to keep the straw bale off the slab and further off the ground (see detail attached). For simplicity of the straw bale construction and for aesthetic reasons, I want to hold that rubble wall line, hence the 24" knee wall on the downhill side.
Oops, second detail didn't attach. Here it is.
Even in NM and the Southwest where we have a hot & dry climate, we see a lot of straw bale walls fail due to moisture coming in. You should install a moisture barrier on the outside and let your wall dry to the inside.
I would omit the spf, but otherwise everything seems fine to me, and good details for hiding the XPS.
Armando - hmm, I have not built one (yet), so my information is second hand, but there has been pretty good luck with lime plaster on bale in the northeast for a number of years. Note that I'm talking lime plaster and not cement stucco - there have been problems in our moist climate with cement stucco due to the tendency to crack and admit bulk water. I continue to look for more straw bale experiences, so i'd be interested to hear more.
Matthew - well, my first pass (as shown in the detail) is xps & cellulose - spf was an alternate to try to solve the condensation issue, perhaps uneccessarily
Is the stud wall structural? If it is not holding up the hay bales why not just go all foam? A PT cleat on the floor to attach whatever the surface you are plastering. Could carry it all the way to meet the underfloor foam and improve the perimeter insulation
Is that post really loading thru the floor and foam to the foundation?
I think I have to consider the stud wall structural - I would not want to try to build a cantilever off of a rubble masonry wall. At one point, I had considered thickening the rubbble wall to the back face of the bale, and building out a wainscot to insulate it, which could be built as you note.
And no, at the post, the bond beam widens and the insulation is omitted for direct load transfer. Bit of a tricky insulation continuity detail there, but should work.
I could not disagree with Armondo more, and I have dozens of straw bale homes behind me, with no failures. Lime plaster can be much more water resistant than tar paper. @ Keith, I think he wants something to attach to, attaching to foam sucks, and relies on adhesives which do fail. DO not question your details here Drat, they are great.
At 6 inches and 2 feet you would not have to attach to foam
Depending on the spacing of the timber posts you could hang a small 'beam' from post to post to carry the possible weight of the bales and you would eliminate bridging and improve the slab perimeter insulation.
I doubt his original plan would be a problem, but he was asking for alternatives
I would suggest keeping any building heated to some degree as the temp swings cause condensation on everything. In addition, with radiant heat it will take at least a day to normalize, so I would fire the slab at least once a day on a timer.
Drat,
I may have missed a point, but If your goal is to stop thermal bridging at the exterior of the stud wall while maintaing vapor permanence... Consider replacing the XPS with a high density Rock Wool Panel like Roxul Rock board 80 or greater. Perhaps place a high quality and vapor open WRB on the face of the stud wall, and then the Roxul over that. This should deal with the issues of both liquid and vapor water if your plan is to be diffusion open and can allow the gap behind the stone to vent properly.
Well now that's interesting, spanning post to post to support the bales. My posts are probably too widely spaced, but I like the idea in principal and will look into it.
I've have swung back and forth on the thermal mass in this building - I want it for the passive solar storage and its simplicity (slab is the finish floor, plaster is the interior and exterior finish), but I worry about it for the long cycle time of bringing it up to temperature. I'm now planning to rely on a wood stove for heat and have been thinking that at most, I would lay the pex (for future flexibility) and see how things go. Based on some things I've read, I wonder if that system would ever be necessary in a tight & well insulated house - and perhaps even the pex is just a waste of money and effort. Still noodling on that one.
And the condensation issue of course goes back to my original posting. Thinking about it, I'm wondering how much of an issue that is? I have no practical experience with it (growing up, our 'camp' was summer only), but if I'm directly exhausting at the primary humidity sources (bath and stove), it may be less of a load that I'm thinking that it will be. I don't have the wherewithal to estimate how much a weekend of activity will raise the humidity inside, nor how the temperature / RH chart will compare once I shut everything down. That'd be an interesting thing to see - I assume that the temperature would would drop faster than the RH, but I assume that the RH will continue to drop until it reaches outdoor levels (assuming i don't have a lot of plants), whereas temperature would level out due to the solar. Wow, Joe Ls. could do a whole bunch of charts on this one.
But now I'm just musing. Thanks for the input so far. More than I expected from a holiday weekend.
Albert, yes that was my original point, and that makes a lot of sense to me. I don't have experience with rockwool on the outside of the WRB (I think of it as firestopping and curtainwall insulation (giving away my background here) - is that an appropriate application? I would have guessed that it would need to be inboard of the WRB.
Drat,
Check in with your local supplier. Rock wool (to me) is an amazing material. In the higher densities, It's really unaffected by water. This means that placing a WRB in front of it... is redundant for liquid water behind a rainscreen. The WRB is there to protect the wood products not the rockwool and is much easier to place at the 2x4 wall and detail correctly. At R4.5/inch, inert to water and bugs, while still vapor diffusive, it may be a good choice for this application.
Good luck!
Drat,
Q. "My first pass at this was (inside-out) : lime plaster on lath, on studs insulated with cellulose, 1-2" rigid XPS, airspace for weeps, rubble masonry. But I'm concerned that I've built a condensation surface at the XPS."
A. You won't have any condensation problems if you use at least R-7.5 of XPS for a 2x4 wall or at least R-11.25 of XPS for a 2x6 wall. There is more information on this issue here: Calculating the Minimum Thickness of Rigid Foam Sheathing.
Following on from Martin's comment, if this advice results in greater thickness of the insulated part of the kneewall assembly you seem to have room on your footing to move the rubble wall outward by at least 2" and you could cantilever it somewhat further if necessary. This would result in a water table detail at the junction with the straw bale/plaster above. This should not be a problem to execute successfully: I assume the lime plaster is protected by a wide roof overhang. And by giving additional expression to the rubble wall and a stronger sense of grounding to the cabin, to my taste at least this slight projection of the stone would actually be a visual benefit.
Thanks Martin. As shown, I have 2" of xps (R10) on 4" studs, so I meet the guidelines and code.
BUT, the guidelines and code are written under the assumption that temperatures will be maintained in the house, keeping the interior of the xps above dew point. As the house cools (even if I don't let it go completely cold, I'll surely let it get down into the 50s), that is going to change the formula and I expect that the xps would at some point fall below the dewpoint. (see my post #14 for my further musings on this issue). That is where I think the problem is.
I suppose I should've titled this posting 'vapor accumulation in a weekend cabin" to clarify my issue, but now that I've written those words, someone will at least find it by searching...
That said, my current preference is for Albert's recommendation for rock wool in lieu of xps - solves the thermal bridging while maintaining vapor permeability.
James, I was avoiding the water table for some reason that I can no longer recall (I agree that it would look better, and i've got 2' roof overhangs not too far overhead). There's a code requirement for 2" footing projection (not sure I understand it), so that may be how I got to where I am.
I may well build that water table anyway - 9" fees a bit skimpy for the thickness of that wall.
Drat,
Cooler interior temperatures shouldn't increase the chance of condensation, since cool air tends to be dry. If the indoor temperature is the same as the outdoor temperature, and the surfaces are also at the outdoor temperature, and the RH is uniform, there will be no condensation.
The risk of condensation increases as interior temperature rises, since hot air can hold more moisture.
The problem with an unheated space is the temperature swings. Everything metal in an unheated space rusts in the winter. The temperature rises during the day and the air can hold more moisture, but all the objects with thermal mass stay cold, and moisture condenses on them.
I imagine in a well insulated space it is not as big an issue as the swings will be smaller, but that is why I suggested keeping a minimum temp, I would say at least 40 degrees[probably not an issue anyway]. Of course it has little to do with moisture drive through the wall, just moisture in general.
Uninhabited houses do get mildew problems.
I still think more insulation is a good idea in that spot.
And whatever I do, I think I should avoid buying a humidifier. :)
Martin,
But cool air is only dry because that air went up over the rockies (got colder) and snowed all the moisture out.
If I'm at my cabin for a cold weekend, I'll build a fire (lowering the RH dramatically as I do so). My skin will crack on that first day because it's so darn dry inside, but I will breath, cook, dry my snow soaked clothing, and take a nice warm shower (and raise the RH to a more comfortable level as I do so). Let's say I'm at 68d and 30% RH at the end of the weekend. Once I kill the fire and the house starts to cool, the RH will gradually start to rise unless the moisture dissipates at the same rate. Assuming that I build a tight house and limit air movement (and if I don't, I have bigger problems), I have to rely on vapor diffusion through the building construction to remove the moisture, don’t I? Are you perhaps thinking that there’s a ventilation solution here? Assuming I’m not venting the vapor, it's a race between r-value and perm rate.
If it's 10d out and 68d in when I leave, that puts the interior surface of the xps at 40d, comfortably above the dewpoint of 35d. But even if I were to hold a minimum temperature of 40d inside as keith recommends (makes a lot of sense to me, and I’d like to think I can hold that with the passive solar), the interior surface of the xps will drop to 25d. The question is whether the humidity drive from inside to outside will have kept pace - otherwise, the xps will hit dewpoint and start to accumulate moisture (likely as frost). Whether that frost melts and evaporates the next weekend or melts and accumulates from one weekend to the next, I think I’ll have a problem in the wall.
I know I originally asked this as a question, but the more I think about it and write about it, the more I’m convinced that it’s an issue. More insulation helps (match the performance of that wall to the straw bale so it cycles the same), but I’m now convinced that the materials making up the wall assembly need to have a similar permeability.
Drat,
As warm air cools, condensation will form on surfaces that are colder than the air temperature. After you leave your weekend home on Sunday afteroon, the surfaces in your house are likely to be warmer than the air temperature as the interior air cools, not colder than the air temperature.
The danger of condensation is more likely to occur later -- when warm, humid exterior air enters your home through cracks after a cold spell that has cooled your furniture, tiles, and interior masonry. This is the scenario that Keith Gustafon accurately raised. However, this scenario has nothing to do with the design of your wall assembly.
Well, I'm still not sure I buy the air colder than the surface argument, but I see your point enough to stop arguing it. But I am planning to switch xps for mineral fiber, because I think it'll make me sleep better.