It's amazing that already 3 months have passed since I moved into the house Oct 25th. It took less than a year to design and build the house. My thanks to the designer,
Dora Coates of Dovecote Design and builder,
Tim Yandow and his many capable sub-contractors.
The view out the south windows since completion has changed from barren trees to a white scene for the last two months with temperatures mostly below freezing, making for great cross country skiing and house heating performance observation.
Living In It
It took a few days to adapt to the new internal environment:
The quiet interior - no blowing air from air ducts every time the heat comes on; barely audible rain and howling wind; no perceived cold spots; the outdoor nearly present indoors due to the large southern windows; the lively acoustics largely due to the linoleum floor; the warm feeling from the adobe colored walls and cinnamon swirl colored floor; the rock-stable and very slowly decreasing basement temperature as we headed toward the minimum average temperature for the year; the operation of the radiant floor water reservoir for the instant-on electric water heaters..... Some of these observations and more will be further explored in what follows.
All in all, I've gotten well adapted to the new house; it's become a part of me and we take good care of each other. I strive to keep the air temperature at 65 degF which means that the floor temperature, my only heat source, has to be regulated at a slightly higher temperature depending on the outside temperature. Typically, at around 10 degF outside the floor temperature needs to be 6 degrees higher than the desired air temperature. No more cold feet.
Heating Energy Demand
Since all of the energy going into the house is electrical which is turned to heat without loss and since heat flow out of the house is largely through the building envelope with some exiting via the wastewater and heat recovery ventilator (HRV), it can be said approximately, that all my electrical consumption goes to heat the house. Extensive daily data taking has shown that my house heating needs to keep it at 65 degF is around 750 W/degree-day. This means that for a 25 degF day, for example, 40 x 750 or 30 kWh of energy is needed. On an annual basis for our average 7446 degree-days/year in this part of Vermont, the house is now estimated to need 5584 kWh/yr which is better than the 6500 kWh from the original heat-loss calculation for the house from week 14. Compare this to the 2008 average annual US residential utility customer consumption of 11,040 kWh (EPA).
Humidity and HRV Performance
The heat recovery ventilator's purpose is to bring in fresh air from the outside and expel the products of respiration (CO2, moisture, and other smells that may accumulate from cooking, etc.) and to do so by recovering heat from the warm stale air and add it to the fresh air. It does this pretty well in winter, dehumidifying the indoor air, especially on really cold days. For example on a 20 degree day with the outside humidity at 41%, the air coming in will be 55 degrees at 33% humidity. The HRV has continuous low and high speed modes as well as a 50% duty cycle low speed mode. I use the high speed mode mostly while cooking something odiferous. The other modes are used mostly on sunny or warmer days.
House Heating and Cooling Rates
The sun at it's maximum angle of 25 degrees in the middle of winter certainly still has a lot of energy to heat the house. Unfortunately, this winter has been less sunny than usual.
On a recent sunny day with the outside temperature in the teens and falling to negative numbers, the indoor air temperature rose from 61 degrees and stayed a comfortable 65 degrees for six hours with the heat turned off the previous night. The floor temperature, however, slowly decreased in the 36 hours after the heat was turned off, from 69 to 63 degrees, being cooled in part by the colder basement (52 degF). After 36 hours it took 60 kWh and six hours to get the floor temperature back to normal. The air temperature took somewhat longer to stabilize as all of the furniture and walls had to heated.
On warmer sunny days it can get too warm inside and the HRV needs to be run continuously or the windows opened on warmer days. More data is needed in milder weather.
One might wonder how the temperature of the house's concrete wall varies with outdoor temperature variations, considering that the walls have an R30 outside insulation and an R10 inside insulation. A test hole was bored to the concrete wall on the north side of the house for temperature monitoring. On days of stable outside temperature, the concrete wall was observed to be cooler by 25% of the difference between inside and outside temperatures, as expected. The thermal mass of the concrete wall stores heat much like a reservoir stores water. The rate at which the heat reservoir fills is a function of its capacity and net flow rates of heat, being limited by the Rvalues of the inside and outside insulation. The capacity and Rvalues define a time constant for the wall for a sudden change of temperature. This time constant has been calculated as 9 hours. It has not been verified by the sparse data taken to date in the presence of quicker varying external temperatures.
Blower Door Test Results
Efficiency Vermont personnel came by Feb. 4 to inspect the house and perform a Blower Door test. This test is performed by installing a temporary "door" with a calibrated fan into the existing entry door way. The inside air is exhausted until a pressure difference of 50 Pascals (metric!; 1 lb/ft2) is achieved and then the air flow rate is measured via the Bernoulli principle. This air is the sum total of all air leaking into the house via any paths that may exist. On a cold day one can walk around and feel for cold air coming from windows, electrical boxes, and floor boards.
The house tested at 400 cubic feet per minute which computes to 3/8 of an air change an hour (ACH) for my house volume (not considering the basement). According to a
good reference on this topic less than 5 ACH is considered a tight house requiring active removal of humidity, combustion, and respiratory gases. And that is my case.
Stove Performance
The Beeeutiful (and dear) wood burning parlor stove has not been used regularly for lack of wood and for not messing with the heat data being taken.
It heats plenty in the well insulated house. Two hours is enough to raise the air temperature 4 degF. On a freezing day, the heat stored in the stove's soapstone and house interior will take six hours to return to the original temperature.
The firebox is a little small, making it ideal for smaller pieces of wood. It's not easy to start the fire in such a small space, however. The curved glass design acts as a heat lens concentrating reflected energy into the center of the firebox enabling a single piece of wood to be burned successfully, something most other stoves cannot do so easily. Also, one must be really careful opening the front door as rapid opening will draw the flame and dust out the opening. Shutting off the external air intake just before opening the door slowly seems to help. It's a nice parlor stove though not so practical for 24 hour heating.
Vermont law requires an outside air supply for all new house construction. This stove was designed to accommodate that requirement with a fresh air adapter. The manufacturer of the adapter was of no help in getting the dimensions of the device, requiring a purchase first. The device would not fit and had to be adapted. Fortunately Metalworks in Burlington was a very capable place for custom modifications. Another problem yet to be solved is condensation on the non-insulated aluminum fresh air supply ducting in the basement on very cold days.
Root Cellar Performance
As a reminder, the root cellar is the space below the shed. It has a nearly 10ft ceiling with a gravel floor to the foundation soil. The upper part of the ceiling is approximate 2 ft. above the outside soil and thus exposed to daily temperature and wind variation as there is currently no insulation on the cellar ceiling/shed floor and the ceiling has not been thoroughly air leak sealed. Long term soil temperature variation are moderated by the 4" foam insulation on the concrete walls.
When the external temperature drops to single digits, the root cellar temperature has been observed to drop below freezing requiring removal of fruits and roots. For most of the winter it has hovered around 35 degF. A two inch reflective polyisocyanurate ceiling insulation addition is being planned.
Water Usage.
I was surprised to find that the average per capita indoor water usage in the USA is 69.3 gal/day! Having a water meter allows one to observe one's water usage and maybe adapt to less consumption if one cares to. I was curios and found out that my one person household uses the following amount of water:
Toilet (5 flushes/day @ 1.3 gal/flush) = 6.5 gal/day
Laundry = 12 gal/wash
Short shower (wet, lather, rinse) = 1.3 gal
Full shower (as above with water running the full time) = 11.5 gal
One-person daily dishes hand washed and dried = 0.6 gal/day
Measured average daily use =12.8 gal/day.
So if I were to flush with my high mineral content community well water and use soft rain water for the other uses, then my annual consumption of soft water might be around 2300 gal/yr or 307 cuft/yr. On my 1277 sqft roof this would require on average a 2.88 in rainfall capture for the year. Considering that the historical average monthly rainfall is 2.9"/month, where's the problem here? Frozen gutters in winter? Three months or 600 gallons of storage?
Hot Water Performance
This house uses two Stiebel-Eltron instant-on electric water heaters, one for the shower and wash machine, the other for the kitchen. Cold water entering the house is 45 degF in winter and gets preheated to 65-70 degF by the radiant floor in a dedicated water loop. This pre-warmed water is what then feeds into the heaters. To achieve temperatures greater than 105 degF the flow rates have to be limited by flow restrictors. This works fine in the kitchen but not yet in the shower, largely because of the lack of a proper control valve for low flow rates. Still looking....
Attic Performance
The attic design evolved after much discussion about venting and moisture infiltration from the non-impervious ceiling typically installed in houses. Traditional thinking assumes that moisture will, nay should, escape via the ceiling into the attic to avoid moisture buildup within the house. Once in the attic the moisture must be vented requiring soffit vents for air entering and ridge vents for air leaving. This natural convection flow, assuming there's heat to drive it, should vent the humidity in the attic. The alternative design based on the presence of a heat recovery ventilator does not depend on attic venting. It uses a ceiling impervious to moisture thus avoiding the need to vent the attic rigorously. Thus this house has neither ridge nor soffit vents, instead relying on two gable vents at opposite ends of the roof for attic air equalization with the outside air.
To test the performance, a remote reading temperature/humidity gauge was installed in the attic. Data from this sensor shows that the moisture does not build up in the attic and pretty much tracks the external moisture and temperature in the absence of solar energy impinging on the roof. On sunny days the attic does get warmer (no data yet for summer). All temperature variations in the attic minimally affect the house temperature because of the R75 insulation.
Windows/Doors
The large windows on the south side have certainly brought the outdoors practically inside. It's been enjoyable watching the change of seasons with the abundant winter weather this year. And it allowed me to easily observe the growth of a 5 ft snow wall created by north wind driven snows swirling on the south side. It's also been interesting seeing the moisture condensation patterns when the temperatures fall below 20 degrees. It's become a part of my early morning rituals on those cold mornings to sponge up the condensation water from the bottom of the window frames. Fortunately they are fiberglass and thus less susceptible to mold formation.
The condensation is not a sign of poor windows but the combination of the house's relatively high humidity (45-50%) and their position on the outside of the window well. The fiberglass and wood doors to the outside, sitting flush with the internal wall, do not exhibit condensation as readily as their surfaces are kept warmer from the room's warm air than the deep seated windows.
To illustrate the difference insulating shutters would make to heat loss as judged by condensation, I temporarily installed custom made R10 polyisocyanurate aluminum clad panels on the inside and over the outside of the window frame in the bedroom, the smallest window. As expected, the inside panel kept the inside heat from reaching the window and the window iced up. The outside panel kept the heat from escaping the window by conduction and radiation and minimum condensation was noticed. Is it time to start incorporating external insulating shutters into future window designs not only to cut down on condensation but also night time heat loss?
Floor Insulation Squabble
As a fair amount of heat can be lost from a heated floor to the cold basement below it, floor insulation becomes important. Traditionally, insulation can be blown in between the 12" deep joists or a sheet of aluminized bubble insulation can be stapled to the top of the joists with a 1" air gap. Getting good R-value numbers for these insulation approaches was difficult as achieved values are said to be installation dependent. I don't like uncertainty in these kind of situations and decided to go with (more costly)1-5/8" polyisocyanurate sheet with an aluminum reflective foil on the floor side and a white aluminum foil on the basement side. These sheets were attached to the bottom of the joists, creating a fairly nice looking ceiling for that part of the basement that has radiant floor heating installed for a future room. The rest of the joists had Reflectix aluminized bubble insulation stapled to the top of the joists. The purpose in mixing the types of insulation was to save some money and to evaluate their performances after completion of the house.
The evaluation has been completed to my best ability to do so. As the picture shows, a piece of insulation with a known R-value is placed under the insulation to be evaluated. The temperatures of the floor, the basement, and the surface between the test and the known insulation are then taken. The R-value of the unknown material or structure (flooring, joists, air space, basement ceiling covering) can then be estimated with the following equation:
Ru = Rk(Th-Ti)/(Ti-Tc), where
Ru = unkown R value of material or structure
Rk = the known R value of the reference insulation
Th = is the warm temperature on the material
Ti = is the intermediate temperature at the material/reference insulation interface
Tc = is the cold temperature on the cool side of the reference insulation.
Measured results show the polyisocyanurate sheets with reflective foil, covering the joists and air space gave an equivalent R36, better than expected. The Reflectix aluminized bubble insulation computed to less than R1. I noticed upon installation of the test structure that I could feel the radiation form the warm floor coming right through the insulation and bare joists. Reflectix as installed was a waste of money!!
Construction Observations
One concern I had with concrete floors was their flatness. Even though the concrete contractor had many tools to even the surface of the poured concrete there were noticeable low and high areas. Some areas had to be leveled prior to laying the linoleum flooring. Rigid furniture required shimming or leg adjustments to avoid that rocking experience which I experienced the first night in bed.
The drywall boards were attached directly to the ICF mounting tabs and the resulting wall was only as flat as the ICF walls, which showed some deviation in places (see builder's commentary below). The baseboard trim as a result, being largely inflexible and straight showed noticeable voids between walls and floor.
Habitat Houses Commentary
The two Habitat for Humanity houses that were built down the street and occasionally discussed in prior postings are nearly completed. Efficiency Vermont will install environmental monitors throughout both houses to monitor every energy related action for characterizing typical energy use in such Passiv Houses. My house will be similarly instrumented by Efficiency Vermont thereafter for similar purposes.
What Next?
I look forward to spring and summer to pretty up the outside in a permacultural way and to observe the house behavior through a Vermont summer without air conditioning. The spirit willing, I may report in another posting on my observations come November. Thanks for reading.
Builder's Commentary at my request:
The Insulated Concrete Form Experience: A Contractors View
By Tim Yandow
I was influenced very early on in my career as a home builder and renovator by the growing green building movement, particularly the focus on energy efficiency and carbon neutrality in regard to both materials and energy consumption. So when I was introduced to the concept of Insulated Concrete Form (ICF) construction, I was immediately intrigued and looked forward to an opportunity to use them. I was delighted when Wolfger Schneider selected me and my crew to build his home using ICF’s. It allowed us to get a first hand experience of how ICF’s work and see their potential in the growing business of re-working the building paradigm in order to respond appropriately and responsibly to the challenges we face with Climate Change.
One of the immediate and obvious advantages to building using ICF’s is the building envelope it creates, especially if the ICF walls extend above the basement or crawl space. In conventional framing, there is always the challenge of trying to create as air tight and continuous an insulated envelope with as few breaks or seams as possible, minimizing thermal bridging (conducting cold and heat from one side of an exterior wall to another) and the need to manage thermal bypass issues (air movement through the wall assembly). Since Wolfger’s house uses ICF’s all the way from the footings to the roof trusses, the foam insulation is continuous. The ICF system we used, Quad Lock, also offered a fair amount of flexibility in terms of the level of R-value that one can achieve, all the way from the basic R-28 ( 2 inches of foam on either side of the concrete core) to R-values above 40.
From a building perspective, using ICF’s gives the builder the flexibility of creating their own foundations and wall systems without having to sub-contract the work to a forms crew. All that is needed is a little training from someone who knows what they are doing, a nice cool sunny day and a boat load of concrete.
The wall systems themselves went up fairly easily and were quite fun to erect. Our supplier, Brian Kiniry, was sick of hearing us say “hey these are just like Legos”. I guess when you hear that everyday for a few years it can get wearing. But they really are like Legos which adds to the fun. Given that this was our first ICF house, we knew from the start that it would take us some time to attain our usual level of efficiency and that there would be a lot of problem solving to contend with. Both were true.
We found that figuring out the door and window bucks was challenging, not so much constructing them, but keeping them from shifting during the concrete pour. The instructions we were given were not entirely clear in this regard, so we invented a system which worked fairly well by extending “legs” down either side of the buck to the subfloor, creating a solid base for the bucks to sit so the weight of the concrete would not push them out of plumb and level. This worked great.
One issue which we faced right from the first pour to the last was settling of the forms. In theory, because the ICF blocks are uniform in height and width, the walls should come out to a perfectly even height once the concrete has been poured and smoothed out on top. We encountered variations in the wall heights in the basement by as much as a ½ inch. I am honestly still not sure why. We checked the height of the footings with a transit and found they were within ¼ inch. We found that any small variations from level or plumb tended to amplify as the walls grew taller. When we reached the top of the first floor wall after the final pour, the final height of the wall varied by 1 inch from one end of the building to another along its greatest length. As a result, we had to set the top plates with a transit and shim and subsequently insulate the gaps this created with foam. We found this very time consuming but vital to making sure the roof truss system went up properly.
We found that this was also true in terms of the plumbness of the walls. The higher we went the harder it was to keep everything plumb. I think the staging system played a role in this given that the walls were straightened out with workers standing on staging screwed into the ICF tie network. Even though we checked and rechecked our string lines, we found that these shifted to a degree during the pour. Concrete is heavy stuff. But all in all, I was impressed with how straight the plain of the walls proved to be, even along its longest length of over 40 feet. I imagine that after doing several of these homes, one would get quite efficient at minimizing these discrepancies and assembling the walls more quickly. In terms of labor, we found we needed more time to assemble the walls than the ICF company estimated. Again, that was due to both our inexperience with this type of construction and the fact that we encountered more challenges than we expected. Everything always looks a lot easier on a DVD then it ever does in the field. The issues of wall height and plumbness were by far the two most difficult issues we had to contend with. I have not had the chance to talk with other contractors who have used the Quad Lock system to see what their experience was. But I would like to at some point. In the end, the house looks beautiful and we found that these issues did not effect the finish work very much or the roof system, but if we had not spent the extra time correcting them, I can see that the finish work would have been much more challenging.
In terms of cost, the ICF walls are clearly more expensive and have higher embodied carbon in comparison to a stick framed wall and cellulose insulation which is what we typically use when building super insulated homes. I estimate that the difference in cost given the same R-value is around 20-30%. This is for the exterior wall assembly, not the cost of the entire house. This is true for foundations as well (the difference between a conventional concrete foundation and an ICF foundation although the ICF foundation will achieve a higher R-value per inch because it uses XPS foam rather than blue board for insulation).
Although there were times I got frustrated with all of the problem solving we faced during the wall construction, in the end the project was a great success and I would consider using this system again. There are many ICF systems out there and I can not speak as to the relative differences in construction ease and cost between them. That would make for an interesting study. But I would say the Quad Lock system works fairly well and with more experience, I could see that they will just get easier and more efficient to use. Was I “wowed” by my ICF experience? Not entirely. I honestly feel that, having done both now, using a double stud wall assembly and cellulose insulation vs. ICF’s is a more cost effective and sustainable way to build even though you have to build thicker walls to get the same R-value and you have to be more careful with thermal bypass issues. This is primarily because one is building with wood and cellulose (made from recycled paper) instead of foam and concrete.