Energy-frugal owner-built house on a budget?
I’d like to start a free-for-all discussion about building an owner built energy-efficient home on a budget. I’ve turned over the options to the point of feeling dizzy (more than usual). I used to build homes (some double walled super-insulated, some architect designed and specified, beautiful, but far from energy-efficient) for other people. Now, I design and build furniture for other people and build structures for us. I will build carefully in a way a builders on a budget may not for reasons of profitability.
We live in the woods of Wisconsin near Mpls/St.Paul, in a small 650 sq. ft. passive solar house (dbl glazed low E, argon gas filled), with hydronic heat in a acid-etched slab (off-peak electric micro boiler). I’ll be adding 600 sq. ft. in 2 stories, with a wood stove.
So, where should I put our $? Weighted towards North American made Passiv certified windows? (Being in the woods, we like windows for the view, solar gain and natural light.) Spray Foam? Solar panels?
I’m leaning towards a carefully sealed 2×6 mineral wool wall with sealed 2″ foil-faced foam on the warm side — approx. R34. I am wary of the durability of windows getting U value with suspended heat reflective film between panes, and of the “bugs” that may yet to be found in the products of the new Passiv window players; I’ve seen some of the German made windows and they are EXTREMELY impressive, but we can’t afford them. Of the more affordable Canadian (or American?) made windows, which seem to have the best track record for performance and durability? Any other thoughts given our parameters?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Edward,
Here's my feedback:
1. Window specs -- especially U-factors and SHGC numbers -- matter more than whether or not a window has been certified as a Passivhaus window. I doubt that paying extra for Passivhaus certification is the best use of your window dollars. Many builders of superinsulated homes are happy with triple-glazed windows with fiberglass frames -- most manufacturers are located in Canada -- even if these windows aren't Passivhaus-certified.
2. Like you, I am not confident in the longevity of the plastic films used to make Heat Mirror glazing. If you want triple glazing, I would go with three layers of glass.
3. If you are planning a wall assembly that includes rigid foam, I advise you to install the rigid foam on the exterior of the wall, not the interior. That way the foam will do a much better job of insulating rim joists and partition intersections.
4. Meticulous air sealing is necessary for good thermal performance. These air sealing details aren't a product that you can point to -- but they make all the difference when it comes to comfort and low energy bills.
In a house that size with such low loads, applying the electric boiler money & associated hydronic system components to 3/4-1 ton cold-weather mini-split is probably going to be a better expenditure of funds, and another when you build out the additional 600'.
With 2" of polyiso (I agree with Martin that the exterior is best) and R23s in a 2x6 16" o.c. spacing your wall U-factor will be about 0.04BTU per square foot per degree-F. Assuming the footprint is roughly 20 x 33', with 9' ceiling the gross wall area is ~1300 square feet. Assuming a 15% window/floor ratio and a couple of doors the net wall area is ~1150'. At an outside design temp of -15F, interior of +70F, that's a delta of 85F, and the wall losses will be:
Walls: 85F x U0.04 x 1150= ~3900 BTU/hr
With 100 square feet of U0.30 high-gain windows the window losses come to:
85F x U0.3 x 100= 2550 BTU/hr
A coupla 25 sguare foot U0.25 (R4) insulated doors comes to: 85F x U0.25 x 50'= ~1100BTU/hr
Assuming 650' of your code-min R49 between 16" o.c.joists you'll be at about U0.023 on the attic for an attic loss of:
85F x U0.23 x 650'= 1270 BTU/hr
Add it all up and you have ~9000BTU/hr. If you build tight and use an HRV your infiltration/ventilation losses will be less than 1000 BTU/hr, but a couple of sleeping humans and a refrigerator buy most of that back, so you're really probably closer to 9000BTU/hr than 10,000 BTU/hr. If you sprung for the super-performance windows you'd peel maybe 1500 BTU/hr off that, but that would still be in 3/4 ton mini-split territory. (At some cost & performance break-point you're better off spending the window money on rooftop photovoltaics, and sticking with a high quality air tight code max U-factor but high-gain windows, and maybe a bit less window area.)
A Fujitsu ASU/AOU 9RLS2-H (the -H is important here) or a Mitsubishi MUZ/MSZ-FE12NA would deliver the whole load at -15F. Mitsubishi MUZ/MSZ-FE09NA would come close. Doored off rooms might need a cheap 300-500W electric cove heater for temperature balancing (a room by room heat load would tell you where), but you'd probably be at no more than $5K, installed. (Cove heaters cost about the same as electric baseboard, but are far more comfortable.)
And importantly in higher-R passive solar house, they both deliver high efficiency air conditioning. With passive solar and high-R your active heating season will likely be 5 months out of the year, but you'll also incur a longer cooling season.
And with a peak load of less than 5BTU per square foot the barefoot cush-factor of a radiant floor is also pretty marginal- you'd feel it on the absolute coldest of days, but on an average winter day during the daylight hours, hardly at all (especially if you're using the wood stove.)
A mini-split solution with judicious use of resistance heating for balance would use less than half the power of an electric boiler, but more than a third in your climate. The installed cost of a 3/4 -1 ton mini-split of that ilk is between $2800-4000, in my neighborhood, depending on the particulars. YMMV. While you might get an appropriately sized electric boiler for between $500-1000, most systems would be looking at another $1000 in pumps, manifolds, and controls even before you get down to the cost of the PEX and plumbing labor. With some DIY labor it would AT BEST the hydronic solution would come in at about the same cost of a mini-split solution, but it would use more than twice the power (still $100s more per year, even at nickel a kilowatt hour- off-peak), and you'd still be left with figuring out something else for air conditioning. A 3/4 ton mini-split should be able to handle your peak loads with margin, unless you go nuts on west-facing window area, and build with minimal roof overhang on the south side.
Thanks for your thoughts, Martin and Dana.
Regarding the foam on the outside rather than on the inside... I have read many of the articles on this method in FHB and elsewhere. There will only be 2 partitions in this project and I would add them after the inside had the foam installed. I would probably use an insulated rim (and if I don't, I thought it might be a worthwhile compromise). The exterior detailing for the exterior foam looks futzy... though perhaps only because I've never done it. Also, here, the foam must be quite thick on the exterior to prevent condensation, and on an intuitive level (I know building science may trump my intuition), preventing the moist air from entering the wall from the inside by meticulous sealing makes sense to me. I plan on using under-course cedar shingles for siding, which with very large overhangs and Galvanized steel for the bottom 3 feet where siding takes the harshest beating in this climate, should perform well enough, but might be challenging to install with exterior foam.
Dana, I much appreciate you taking the time to calculate the buildings heating/cooling requirements. However, the hydronic system is already in place in the existing building. When I built 5 years ago, I sized the micro-boiler and manifold etc. for this addition and put a PVC chase under the slab to facilitate the running of more tubing. So, the expense is minimal, especially since I will be doing the work -- and by adding a wood stove, the system will get little use if we are home. Regarding cooling, the house has large overhangs and, being in the woods, is mostly in shade during the summer. Now, we use a small window AC for the 5-10 days a summer we need it. I may add a very small mini-split anyways, though I need to calculate to see if I think it is worth it.
Martin, since my initial post I've gone back and read some of the triple-pane discussions on GBA. I see you used Thermotech windows on your place a few years ago. Given all the options, how did you decide on them? How do you like them after a few years? How do they look?
I hesitate to buy windows before seeing them in the flesh (or the fiberglass). I used Kolbe & Kolbe dbl panes in the original house and made the jambs for all the fixed panes using Cardinal Glass Low-E, argon-gas-filled patio door glass blems (very affordable).
Anyone have experience with Kolbe & Kolbe triple glazed casements? I know the glazing is only 7/8" thick, which is not optimal, but I wonder how much efficiency is lost --- and their factory is not far from here (convenient and smaller transport carbon footprint) and, the K&K dbl glazed, for what they are, seem like pretty good windows.
Thanks for providing such a great forum for these discussions.
Ed
Edward,
One of my Thermotech casements has a bad crank-out mechanism. Stephen Thwaites of Thermotech sent me new hardware parts at no charge, but I haven't gotten around to installing the parts yet. I don't know if I got the parts at no charge because I'm a journalist. (In any case, I certainly had to pay for the windows.) I know for a fact that Thermotech Fiberglass is very slow to respond to customer service issues. The windows look fine.
With 2 x 6 framing 2" of exterior polyiso (R12-R13) is more than sufficient to meet code in your US climate zone 6 location with only a class-III vapor retarder (standard latex paint) on the interior. Code sez anything over R11.25 is good enough- R13 would give you real margin. See:
http://publicecodes.cyberregs.com/icod/irc/2012/icod_irc_2012_7_sec002_par025.htm
By putting the foam on the exterior, the annual average temperature of the wood is warmer, which means it is also drier. Exterior side is DEFINITELY the better place to put it, as long as it's at or above the code minimum R that allows you to let the sheathing dry toward the interior. By leaving it comparatively vapor open to the conditioned interior the assembly becomes much more resilient to incidental bulk-water wetting, and those wetting events become less likely with the foam on the exterior to boot.
Edward,
I am guessing that the original patio door IG's plus the Kolbe windows all came from Arctic Glass?
A shame that they had to close down. I liked that place.
Oberon
Thanks, all:
Martin, what is the SHG coefficient on your Thermotechs? How was delivery handled to Northern Vermont? Do you know what their lead time is?
Dana, I know most builders who use foam (if it isn't spray foam) use it on the outside. It does make sense to me to keep the cold out of the wall assembly from its source, the outside. It also, though, makes sense to me to keep the moisture out of a carefully detailed wall assembly, from its source, the inside. Will you please say more about "incidental water bulking". Also, I am somewhat resigned to the notion that rigid foam is worth using even though It is made of some nasty stuff, but has anyone used Roxul panels instead with good results?
Greg/Oberon, yes, I got both the Cardinal blems and the K&K windows from Arctic Glass. I miss it too, as it was just a 15 minute drive from here. The fixed panes I used for the south side were affordable for us and heat the house from 10am to 6pm on sunny winter days (even when it is below zero) and all day in the transitional seasons -- and let in needed daylight in the summer given we are mostly shaded -- but they are a net loss because of cloudy days and the night. Eventually we hope to have insulated blinds but they are expensive, which encourages me to use triple panes in the first place.
ANYONE have experience with Kolbe triple panes?? If they perform quite well (even if not optimally) and they cost less than some of the others, I am tempted to use them as there would be visual continuity with what we have already built.
Thanks,
Ed
The interior side is almost never the source of liquid moisture (except in poorly built bathrooms) the exterior side usually is, in any climate.. Interior side moisture drives are primarily air-transported convection/exfiltration at leakage point, and vapor diffusion through the material. Keeping the average temp of the structural wood above the dew point of the interior air and keeping the interior side materials comparatively vapor open so that even liquid moisture that finds it's way in can dry toward the interior at a reasonable rate all but eliminates damaging moisture accumulation from interior moisture drives.
Bulk-water wetting occurs in a number of ways- wind-driven rain pressures backing up liquid water behind the window flashing is a really common one, roof leaks are another, incidental locations where the house wrap or felt is improperly lapped yet another. All siding leaks liquid water eventually, under some circumstance, as do all roofs. The ability of the assemblies to dry determines whether those bulk-water incursions quickly turn into an expensive repair, or simply dry without damage.
In general it's easier to maintain long term air-tightness of the assembly using the exterior sheathing as the primary air barrier rather than the interior gypsum. On the interior side the air & vapor barriers inevitably get dinged up with picture-hanger holes, etc, and there is an order of magnitude more penetrations from plumbing & electrical boxes that need to be sealed perfectly on day-1, and still be air tight 30, 50, 80 years thereafter. Air leakage that reaches cold sheathing in winter loads it up with moisture. But air leakage that reaches 40F average temp sheathing, not so much. The assembly is just much more resilient to the inevitable imperfections with the foam on the exterior.