Design Assessment of Super-Insulated Houses in the Keewatin District of the Northwest Territories
In 1981, seven two-storey demonstration houses were designed and constructed in the Keewatin District of the Northwest Territories. They were designed to be highly energy-efficient and suited to the northern climate and lifestyle, and each house was designed to be self-sufficient for at least two weeks if the community power supply failed. They were built under the authority of the Northwest Territories Housing Corporation (NWTHC), with financing from Canada Mortgage and Housing Corporation (CMHC).
The energy performance and fuel consumption of four of these demonstration houses were measured to compare them with eight single-storey control houses known as Woolfenden units. The conventionally designed Woolfenden units have the same floor area as the demonstration houses and were also constructed in 1980 and 1981. They are located in the same communities as the demonstration houses.
Housing Problems in the North
People living in the North face specific problems that affect housing operation and performance, such as:
Housing requirements specific to the North include extensive storage space for bulk foods; adequate protected space for storage and repair of hunting and fishing equipment and vehicles such as snowmobiles; and extra living space for long-term visiting.
Description of Houses
The demonstration units are two-storey, detached houses as illustrated in Figure 1. Compared with one-storey houses of equal floor space, two-storey houses expose less surface area to the elements, make better use of interior space heaters because hot air from the first floor naturally rises to the second, and provide greater structural rigidity.
The ceiling, floor and walls are heavily insulated to minimize heat losses. (Ceiling, RSI 10.6 (R60), Floor, RSI 8.8 (R50), and Walls, RSI 7.0 (R40)). Careful installation of air and vapour barriers was a priority, so that heat loss and moisture damage caused by the movement of water vapour into the insulation and structure would be minimized. Thick wall sections of 27.9 cm (11 inches) were designed to accommodate the large amount of insulation used. The thick walls also permitted the designers to construct rigid joints at floors and roof. This prevented separation and air infiltration at connections due to foundation movement. The foundation is a continuous-span laminated beam system which rests on a gravel pad.
An unheated sun porch, with large window area, was built across the south side (front) of the house to serve as a wind buffer. It protects the main door while providing extra storage. This porch was expected to provide solar heat gain to the house during the spring and fall.Most of the windows in the house are also on the south side to increase solar gain.
The layout of the house is shown in Figure 2. It includes a living room, a kitchen and dining area, and a staircase up to the bedrooms located on the upper level. The bathroom is on the landing. Two of the three children's bedrooms use moveable closets to divide the space. Only the master bedroom has a door.
Environmental Control and Mechanical Systems
An oil-fired and gravity fed space heater was chosen to reduce dependence on community electricity, which is subject to outages and is also more expensive than oil. Heated air is distributed throughout the house by natural convection. This can be assisted by a manually controlled fan which draws warm air from the top of the second floor down through a duct to the first floor ceiling.
A heat exchanger was included in the design to remove warm, stale air from the bathroom and kitchen while heating the incoming cold fresh air. The fresh air intake is located in the sun porch.
Hot water is provided by an oil-fired water heater. Both cold and hot water work on a gravity-feed system with the tanks located above the bathroom ceiling. The gravity-feed system was designed to ensure operation during power outages.
Discussions and Findings
The review of this demonstration project included interviews with the families who lived in two of the houses. Interviews were also conducted with the site supervisor for one of the houses and the designer of the home. Their comments are incorporated in the following conclusions and recommendations.
On Site Construction
Difficulty was encountered with the erection of the full height wall trusses because of their flimsiness, the presence of winds, and the absence of scaffolding (which would be expensive to provide). Plans were in metric but the materials were provided mainly with imperial measures, which caused a lot of confusion on site. The grade beams were to be field-fabricated but were delivered as glue-laminated components at a considerable saving in time and money. The site supervisor felt that the (1983) total turnkey costs were in the order of $110-$120 per square foot, or $1,180-$1,290 per square metre.
Comparison of Demonstration Houses to Control Houses
Envelope performance results indicated that the demonstration houses had an average 14% less heat loss than the control houses. However, over a 12-month period the average space heating demands were 151 GJ for the demonstration house and 152 GJ for the control home. This loss in performance can be mainly attributed to low space heater efficiencies.
Furnace efficiency tests confirmed that the oil-drip fire-pot space heaters that were used were only 40-70% efficient compared to the 70-80% efficiency measured from the Olsen oil-fired forced-air furnaces in the Woolfendens.
The demonstration houses proved to be about twice as airtight as the control houses. Indoor relative humidities in both the demonstration and control houses were between 20 and 30%. The houses were also monitored for the severity of moisture accumulation in walls and ceilings. Saturated walls and insulation were detected during tests conducted in December 1981 and February 1982. However, further readings taken in September 1982 indicated that the walls and insulation were generally dry; the building shells appeared to be dry and free from any moisture problems.
Both sets of houses were monitored for foundation stability. The control houses, which are supported by a series of 38 x 140 mm (2 x 6 in.) wood pads, exhibited movements of 10 mm (0.4 in.) or less with no structural damage. The majority of the demonstration houses also recorded movements of 10 mm (0.4 in.) or less. The single exception exhibited movements of between 7 and 30 mm (0.3 and 1.2 in.). This was due to several unique circumstances, such as re-orientation of the house after construction.
Design Features - Demonstration Houses
The two-storey design was well received. However, the design can be improved by addressing the following:
The clear width of the staircase should be increased to a minimum of 850 mm (34 in.); ease of movement and safety are more critical in two-storey houses.
The moveable bedroom closets which were intended as room dividers did not provide enough privacy because of low height. Residents preferred that all of the bedrooms be separated by full partitions and have their own doors.
There was a lack of consensus about the adequacy of storage space. One family indicated storage space was adequate while the other family had erected a lean-to.
The sunporch experienced temperatures only 10 °C (18 °F) warmer than the exterior in winter and in summer it reached temperatures up to 49 °C (120 °F). It is unheated, uninsulated on five of six faces and has a relatively large glass area. The grilles that admit outside air into the sun porch for the heat exchanger intake negate any advantage the porch has as an energy saving device in the winter. The lack of a heat sink or ventilation system makes the space uncomfortable in summer. The sun porch was effective as a wind buffer, minimizing air infiltration at the main door. Despite this, occupants experienced some icing of the main door when the nearby porch door was not kept fully closed.
The sun porch could be retained in its present south orientation if insulated for winter use, and with adequate ventilation for summer. As an alternative, the sun porch could be moved to the northern exposure, where it would not heat up in summer but would still provide additional storage space. A windporch for the main door should then be provided.
For its own protection, the air-vapour barrier should be placed deeper into the wall as part of a double-wall, standoff-wall or strapped-wall truss system.
A weather barrier should be placed under the external sheathing to minimize wind infiltration.
The floors were reported to be cold and subject to air leakage. To solve these problems, floor construction should be stressed-skin panels, topped by a 64 mm (22 in.) layer of insulation placed between 38 x 64 mm (2 x 3 in.) sub-floor support members.
Hardboard should not be used as an interior finish; it was too hard for the families to mount hooks or hang pictures.
Environmental Control and Mechanical Systems
The space heaters overheated and required constant adjustment. They burned continuously, drawing outdoor air into the house through the dryer vent, both ports of the heat exchanger and other openings, while expelling indoor air up the flue with a draft pressure three to six times that of a normal household furnace. It is suggested that these units be replaced with a hot water heater feeding a hydronic system.
In areas where wood is available, an airtight wood stove could be installed in the main living area as a backup or alternative system.
Originally the oil tank was located inside the home, the thought being the oil would have to be warm to flow in the gravity feed system. The smell of oil permeated the house, and when the tanks were filled, the cold oil overflowed when heated to room temperature. The tanks were relocated to the exterior without any reported problems.
Because the air-to-air heat exchangers were noisy and created cool drafts, they were used very little or not at all by the residents. The exchangers should be replaced with quieter, better quality units now available on the market.
Because of occupants' complaints of a lack of water pressure, a pressurized system has been installed. The water supply tank should remain on the upper floor so that a backup gravity feed can be provided during power outages.
The demonstration houses built in Keewatin were generally well received and performed quite well. When designing similar housing for the North, improvements should be made to the construction system, design elements such as stair width and wall dividers, sun porch, and the gravity water and fuel oil systems.
Research Report: Design Assessment of Super-Insulated Demonstration Houses Built in the Keewatin, Northwest Territories and Preliminary Designs for Follow-On Units, September 1983.
Research Report: Monitoring New Keewatin Housing: Final Report, January 1983.
A full report on this research project is available from the Canadian Housing Information Centre.
Contacts | Northern Related Links | Bulletin Board
Upcoming Events | Jobs
Please send any comments or suggestions to AHFC.