Passive House Passes Test in Maine
In August 2011, GO Logic, LLC, a firm headed by architect Matthew O’Malia and general contractor Alan Gibson, finished construction on the first Passive House (PH) dormitory in the United States. Located on the campus of Unity College, near Belfast, Maine, the 2,200 ft2 student residence hall, dubbed TerraHaus, houses ten students in six bedrooms. The residence comes complete with a combined kitchen and dining area and other smaller shared spaces.
This far north, where temperatures average 20ºF in January, meeting the PH space-heating energy requirement of 1.4 kWh per square foot per year means no fooling around when it comes to the building’s envelope. The superinsulated slab includes 8 inches of rigid expanded polystyrene insulation underneath it, yielding a total R-value of 36. The R-50 exterior walls were built using conventional 2 x 4 framed stud walls filled with blown-in fiberglass that are surrounded by 8-inch structural insulated panels. The roof assembly, with a total R-value of 84–100, includes trusses filled with 24–30 inches of blown-in cellulose topped by a standing-seam metal roof. And then there are the triple-pane windows from Germany, the continuous air barrier layer, and the careful air sealing to reach an airtightness level of 0.58 ACH50.
The design choices were guided, as is the case with all PH projects, by modeling the proposed building using the Passive House Planning Package (PHPP), a building-energy-modeling software program. Two years after students first unpacked their bags in the dorm rooms, I checked in with Gibson to see how the dorm’s energy use predicted by the PHPP compared to the actual energy use. In early 2012 Gibson had installed an eMonitor system in TerraHaus to track temperature, relative humidity (RH), and indoor air quality in different rooms. The electric circuits are also being monitored to capture the students’ electrical use.
Modeled Versus Actual Use
The PHPP-modeled annual heating demand was just over 2,000 kWh—not much, as according to the PHPP, 54% of the warmth needed was expected to be generated passively by the sun streaming in through the large, south-facing windows and by all the cooking, showering, and other activities of the college students. Should more heat be called for, there are two supplemental heating systems. A Daikin air source heat pump was installed in the main living space to supply heating or cooling that could then be circulated by the heat recovery ventilator (HRV) and by natural convection to the upper level through the open stairway. To further help with circulation, GO Logic also installed a thermostatically controlled exhaust fan at the top of the stairwell to draw solar-heated air from there, sending it through ductwork and out to the kitchen ceiling on the north wall. To allow for more occupant control of heating, GO Logic put electric baseboard heating in each bedroom and bathroom; each baseboard unit is controlled by an individual thermostat.
As it turns out, those individual thermostats have been more popular than anyone expected, with some students’ bedrooms kept as cool as the low 60s and others as warm as 82ºF on a regular basis. In spite of this, actual heating energy use has been less than predicted. For example, from October 2012 to October 2013, the energy used to heat and cool the building totaled 1,744 kWh. “We are seeing that actual heating use is less than predicted in all of our buildings,” says Gibson, who is monitoring several other completed PH buildings. On the other hand, Gibson says that cooling energy use tends to be higher—although still not very much higher, given where he builds. In his experience, the PHPP doesn’t seem to be that great at modeling cooling energy loads.
Gibson was thinking that hot-water use might be much higher than the PHPP had forecast, but that hasn’t turned out to be the case. The first year, predicted and actual water-heating energy use were pretty close. The second year, hot-water use was a little higher than the year before and even a little higher than what the PHPP showed.
The total site energy use predicted by the PHPP was 6,096 kWh per year. Gibson says the monitored energy use for one year was 7,568 kWh, and the difference is mostly due to household electricity use and the kitchen appliances. Not too surprisingly, the students are using much more energy to power their laptops, TVs, and lights than the PHPP had predicted. As Gibson says, “Occupant behavior drives these loads more than what you can predict or control.”
The results at TerraHaus are consistent with what Gibson is seeing in the other projects he is monitoring: Generally, the actual heating energy use is less than the modeled use; the modeled water-heating energy is on target; and the miscellaneous electric loads suck up more power than predicted, which offsets some of the heating demand. On balance, says Gibson, it all works out. “In general, I’ve been really amazed and happy with how close the PHPP predicted energy use and monitored and actual results are,” he says.
So have there been any surprises? It turns out the RH has not been quite as predictable. In the TerraHaus, an HRV was used to ventilate the building, even though an energy recovery ventilator (ERV) is often specified in a mixed-humid climate, because it conserves humidity in the house during the dry winter months and helps to exclude humidity during the more-humid summer months. With ten students cooking and showering in a tight space, GO Logic was concerned that an ERV might recirculate too much humidity back into the space, promoting the growth of mildew and mold. Surprisingly, it turns out that the HRV is bringing the RH down to less than 30% for most of the winter—not optimally comfortable, but better than mold problems.
The heat pump hasn’t been working out ideally either, but that’s partly due to that uncontrollable occupant behavior. The heat pump was installed in the common room, which is the space that is most heated by passive solar. Between that free heat and the body heat when the room is occupied, the room gets warm without any additional heating, so the students don’t turn on the heat pump. But then the perimeter rooms don’t get warm, leading some students to crank up the baseboard heaters. If the students were using the heat pump more often, and if it were supplying 75% of the heat load as had been anticipated, heating energy use would have been more in line with the original, much lower expected use. If that had been the case, because the heat pump has a rated efficiency of 275%, it would have taken only 1,144 kWh to heat the building, says Gibson—which is the amount shown in Table 1.
Gibson says that as far as he knows, a central heat pump has worked out fine in other buildings, but he hasn’t chosen to test out that strategy in the cohousing community that GO Logic has been building in Belfast, Maine. The Belfast Cohousing and Ecovillage consists of 36 dwellings—duplexes, triplexes, and a quad—that vary in size from 500 to 1,500 square feet. Construction started in October 2011, and the entire community should be built by mid-2014. The heating demand in these buildings is low enough that the cost of a heat pump just isn’t justified. All supplemental heating—other than that supplied by passive solar and internal heat gains—is being met with electric baseboard heating.
Since building the TerraHaus, GO Logic has beefed up its standard PH walls a bit. Originally, the interior walls in its double-wall construction were built using 2 x 4s. Now GO Logic has switched to 2 x 6 internal walls, increasing its overall wall R-value by R-7. The firm has also changed its foundations a little and the foam insulation that it uses to maximize R-values. These changes have helped GO Logic to attain the PH standard’s annual heating demand target, even in buildings that don't have perfect solar orientation. “The PHPP is always helpful in informing us what we need to do to get to the PH-certified level,” says Gibson.
The 2 x 6 walls have other benefits as well. Namely, they have made the structural engineer happy, especially about those walls where there are large picture windows or many plumbing penetrations. In addition, as the labor is the same for either size, the only additional costs are for the bigger lumber and the added insulation. Overall, the payback for these improvements is pretty quick.
Gibson himself is shortly going to be experiencing all the benefits of PH living; he is moving into one of the Ecovillage dwellings. At press time, his house was almost finished, and already he is delighted with its performance. “I will go in in the morning when it is 20ºF outside, and it is 60ºF inside even with no one living there. Then I turn on the lights and work for a while, and it gets up to 65ºF.”
Sitting here in Northern California, I’m thinking those Maine folks are a hardy bunch. Gibson reports that the residents of the original GoHome practically never turn the heat on, and a couple living in a replica of that house turn the heat on only briefly in the bathrooms in the morning. The temperature in that house may range from 60ºF on a cold morning to 75ºF on a sunny winter afternoon. Because the solar heat is free, they just prefer to wait a bit for the house to warm up. Maybe it’s not the homeowners that are hardy; rather, it’s their beefy PH envelopes.
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)
While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.