Building performance

Building performance in use

In 2000 a building monitoring program was conducted to measure (every half hour) the indoor temperature and humidity as well as the outside weather (temperature, humidity, wind speed and solar radiation). The thermal performance of the building was also simulated using the ENER-WIN© program (Degelman and Soebarto 1995, Degelman 2000) and Ecotect (Marsh 2000), and the results of the simulation were calibrated to measured data. Operational energy data (which were then used to estimate greenhouse gas emission) were obtained from the utility records and on-site data.

Table 1 shows the building's purchased electricity use. The total electricity use during the first year was 3671 kWh or 13.2 Gj. During this time the photovoltaic panels had not been installed. The electricity was used for lights and appliances. After the photovoltaic panels were installed and operating the purchased electricity used for a 6-month period was 612 kWh.

During winter it was estimated that an average of 12 kg of wood was burned every night for the slow combustion heater. Thus for 85 nights in winter a total of 1020 kg of wood was used. It is estimated that this amount is equivalent to 16.5 Gj of energy.* The LPG consumption for the gas cooking during this first year was 60 kg or 3 Gj.† Therefore, the whole house energy consumption during the first year (electricity, wood and LPG) was 32.7 Gj.

*1 kg of hardwood= 16.2 MJ (Liversidge, 1981)

†1 kg of LPG= 49.6 MJ

Detail of living room showing wood stove and earth walls

Table 1: Electrical Consumption from November 1999 to April 2001

The estimated CO2 gas emission of all fuel energy used in the house during the firrst year of occupation is presented in Table 2. It is interesting to notice that even though the annual space heating energy is more than the electrical energy, the CO2 gas emission from space heating is actually less than the CO2 emitted by the total electricity use. This is because wood was used as the heating source. The estimated total CO2 gas emission from the house is 5.7 ton per year; however, keep in mind that the wood is home grown. Also notice that the purchased electricity used in the second year of occupation has decreased due to the electricity supplied by the photovoltaic panels, which means that the CO2 gas emission from this house in the coming years will also decrease.

Table 2: Estimated CO2 gas emission during the first year

Note: Conversion factors (AGO, 1999):
* 1.02kg of CO2 per kWh of electricity
* 1.8kg of CO2 per kg of wood
* 2.95kg of CO2 per kg of LPG

Table 3 shows the estimated embodied energy of the building envelope, finishes, fittings and wiring. The method to estimate the embodied energy was based on work by Pullen (1995). The estimated total embodied energy, assuming that all the materials were new, is 619,228 MJ or 5GJ/sqm. Because most of the timber used in the building was recycled timber and parts of the floor and brick walls were also from used or recycled materials, the actual embodied energy would actually be lower, which is approximately 526034 MJ or 4.3 GJ / sqm. Looking at this result, the embodied energy of the roof and wall cladding is obviously quite substantial due to use of the metal. The embodied energy from the window / door glazing is also relatively high.

Compared to the energy use in a 'standard' house in South Australia as projected by the Australian Greenhouse Office, the total energy use in this house is lower. The projected energy consumption of a standard house in South Australia for the year 2000 is 41.5 GJ (AGO, 1999; Tables 2 and 35) whereas this building used 32.7 GJ. The AGO projected CO2 gas emissions from a household in South Australia in 2000 is 7.1 ton per year (AGO, 1999; Tables 3 and 35) whereas this building emitted 5.7 ton.

Table 3: Estimated embodied energy (if all materials are new)

Indoor Thermal Comfort

Several monitoring results are presented in the Figures below. As can be seen, the thermal mass effect of the rammed earth wall, earth bermed wall, and concrete floor helps maintain the indoor temperature at a comfortable level especially during cool nights. The temperature difference between inside and outside was between 5 to 8°C. However, it seems that the effectiveness of the mass was reduced by the amount of north facing glass which means that the indoor temperature could reach around 30°C when it was 35°C outside. During the monitoring period no blinds were used even though they were already installed. Using the Ecotect simulation program, it was predicted that the north facing glass, was shaded only partially during late summer to early autumn as well as admitted reflected heat from the ground (Figure 2).

In the winter the results show that the house was quite comfortable all the time although it was more on the cool side of the 'comfort zone'. Without the heater the temperature difference between inside and outside was around 8°C. When used, the heater increased the indoor temperature by 5 to 10°C. It can be seen, however, that the heater was not used every night.

The simulation results show that after sunset the glazing becomes the major heat loss even though during the day the house is quite warm. To reduce heat loss through the glass, the owner has installed sheets of shrink-wrap to create a 'double glazing' effect on the upper level windows.

Two weeks monitoring in January 2000 of the indoor temperature in the living room

Two weeks monitoring in April 2000 of the indoor temperature in the living room

Two weeks monitoring in July 2000 of the indoor temperature in the living room

Fig.2 Solar shading study of the living space for 4 pm on February 21

Benchmarking

The energy performance of the building was also analysed using NatHERS and the result showed that the building only received a 1 star rating. This result was quite surprising, considering that the actual (measured) performance of the house, as shown above, is higher than that of standard houses in the area. This outcome, however, was more of a result of the limitations in the rating tool, as pointed out in Soebarto (2000), therefore should not be used as a benchmarking.

Qualitative

The occupants indicate that they are satisfied with the quality of living in this house. Living in the 'bush' as well as being able to view and feel the outside all the time is an important aspect in this family. According to the occupants the house is quite comfortable during the year. However in the winter the house was rather too drafty and therefore they installed rubber seals around the doors to reduce air infiltration and a sliding insulated door across the passage to prevent warmed air from escaping to the back door and instead move it to the bedrooms and study. Translucent inserts are also installed in the south facing bedroom windows to reduce heat loss. In summer days, although the monitoring results showed a number of warm temperatures in the house, the occupants did not seem to feel them as they were mostly out of the house during that time. In the afternoon the house cooled down quickly if they opened the doors and upper level windows to let the warm air out. If the house was occupied all the time, however, the occupants may have had a different perception and the blinds may have been used to reduce the heat gain.