Inner City Residential Energy Performance

FINAL REPORT

June 2004

Submitted to:

State Energy Research Advisory Committee
Level 19
30 Wakefield Street
Adelaide SA 5000

Submitted by:

Urban Ecology Australia
105 Sturt St.
Adelaide SA 5000
Tel: (08)82126760
Email: urbanec@urbanecology.org.au

AUTHOR:

Monica Oliphant
Consultant and Adj Prof Research University of SA
Tel: (08) 82773357
Email: oliphant@senet.com.au

1. SUMMARY AND RECOMMENDATIONS

Summary

Energy use in the home can be regarded as being affected by 3 major factors:

The relative contribution by each of the above is not well understood and will differ in various climate zones. The impact of behavior, in particular, is difficult to quantify as is the impact of a pleasant, light and airy well built home on comfort and wellbeing.

In the small inner city housing development at Christie Walk an attempt has been made by Dr Paul Downton to tackle all 3 major influences. This project did not set out to prove whether he has been successful in all, but to tackle the smaller issue of whether the innovatively designed and built homes are low energy dwellings and if so how much better are they than the SA average.

Additionally one of the main problems of the SA electricity system is a high peak summer load driven mainly by domestic air-conditioners.

An analysis of the contribution to peak load by the Christie Walk development was undertaken and a comparison made with the results of monitoring 6 homes at the new housing development at Mawson Lakes. The Sustainable Energy Centre of the University of South Australia, Mawson Lakes Campus, provided the data on which the comparison is based. (It is expected that other new housing developments will exhibit similar outcomes, Mawson Lakes was chosen only because data was available.)

Results have been very positive for the Christie Walk development. The homes have an occupancy rate of mainly one or two persons and are virtually all electric. A comparison with the State average for such homes shows that there has been:

Additionally when a comparison with the 6 monitored homes at Mawson Lakes was undertaken it was found that on the peak load day the contribution to peak load averaged 0.4 kW/household at Christie Walk and 4.0 kW/household at Mawson Lakes. (Note that sample sizes are small, however, it is expected that more rigorous work will confirm a large difference.)

Other issues of note are that it is felt that in SA the better indicator to compare house performance is (t CO2) or kWh /capita /year rather than (t CO2) or kWh/capita/m2/y. The reason is that in a previous study the author found that the parameters that significantly impact on household energy use are, number of people in the home, appliance mix and probably behavior. Though area of house is significant in influencing energy use, the significance does not change markedly as the number of rooms in the home increases from 5 ­ 9. By using area as an indicator larger homes artificially appear to perform better than smaller ones.

Finally, it was found that all of the solar water heaters at Christie Walk are being given a daytime electricity boost. This is not only bad for optimising solar output from a SWH but it is also bad from a utility point of view, as the customer is paying off-peak rates during the peak period and also contributing to peak loads.

2. PROJECT DESCRIPTION

Description of the Christie Walk Project

The Christie Walk project in Adelaide’s southwest quarter was designed to test and demonstrate the processes, plans and principles contained in the ‘ecological city’ vision of the non-profit environmental education association Urban Ecology Australia Inc (UEA) founded in 1991.

The design brief was based on energy efficiency, the use of ‘renewables’ and a high overall ecological performance allied to user-participation in the design and development process. It was intended to set parameters for a project able to demonstrate both the physical and organisational aspects of community and ecological development.

Once it is complete, almost every aspect of ecological urban development will be represented in the Christie Walk project and it is already cited as an exemplar of this kind of development, eg. by the Federal Government’s Sustainable Cities Inquiry and in international publications like the Sustainable Urban Development Reader by Wheeler and Beatley, a text-book with international readership in which Christie Walk is one of 24 case studies of ‘real-world sustainable urban planning examples’.

Providing a ‘real-world’ example was a major goal of the protagonists for Christie Walk and a key aspect of this project is its inner-city location. There are very few examples anywhere in the world of substantial inner-city ecological housing development. Christie Walk is situated in the most mixed-use, least wealthy and most culturally diverse part of the City of Adelaide requiring the design to address complex inner-urban contextual demands. That context supplies solutions as well as challenges - transport energy use is minimised by the site’s walkable proximity to all major urban facilities and the closeness of public transport.

The first two stages of the three stage development are on 1,500 sq.m. of the 2,000 sq.m. site and comprise a 14 dwelling co-housing development organised and managed by its residents. A number of housing types are represented, some linked physically and all connected through landscaping that has been designed to be an integral part of the passive climate response of the dwellings. An apartment building is planned for the Sturt Street frontage that will include community facilities for the whole project.

The overall design strategy was, regardless of orientation, to use high internal mass within highly insulated skins (including double glazing) with multiple user-controlled ventilation options and thermal flues. Solar exposure and control was to be varied according to orientation options whilst minimising overshadowing impacts of adjacent structures.

A variety of construction methods are employed in the various buildings including load-bearing autoclaved aerated concrete, poured low-strength concrete (‘earthcrete’), steel framing, and timber-framed strawbale. Non-toxic construction and finishes are used throughout with a policy of avoiding formaldehyde and minimising the use of pvc. All timbers are plantation or recycled. All dwellings have solar hot water, photovoltaic panels are being mounted on pergolas over the apartment block roof garden, sewage is being treated on-site (thanks to a ‘Coast and Clean Seas’ grant) and stormwater is captured for use on site. Landscaping is based on low water use plantings that favour native and indigenous species with pavings, carports and feature elements constructed from recycled materials including bricks, stone, steel and timber retrieved from demolition of the few pre-existing structures on the site.

The final design of dwellings for stages one and two is for a block of four linked three-storey townhouses with full solar orientation, a three storey block of six apartments with east-west orientation, and four standalone two-storey cottages, one of which possesses an attic third storey and is situated on Russell Street. The four linked townhouses and two of the strawbale cottages (including the one with an attic storey) are the objects of study for this research project.

The townhouses possess ‘classic’ orientation to the solar direction with balconies and shading elements that provide seasonal control of solar penetration.

As their northern aspect is impacted strongly by adjacent structures on the neighbouring site, the detached strawbale cottages depart from this format somewhat, each having orientation primarily to the aspect of the shared community garden space. 63 Russell Street has direct street frontage to the east with its main windows for view and solar gain oriented to the west. It has very few and very small windows to the south and limited windows to the north. 65 Russell Street has no windows to the west, limited southern windows, some northern aspect (with a pergola yet to be constructed over the main northern exposure) and its main windows for view and solar gain to the east.

It is a central precept of ecological design that both its form and processes need to evolve in the context of use and application of principle in the real world. Thus Christie Walk is as much an experiment in community processes and urbanism as it is in making environmentally responsible buildings - its interaction with neighbours and the developing fabric of the city are part of its community enterprise.

The project was intentionally designed to try and encourage and facilitate the idea of the ‘urban village’ and the individual building orientations are intended to address community space and passive surveillance requirements as much as the imperatives for passive climate response. Within another 12 months the extensive plantings that are part of the finished landscaping will be in place with the expectation that the performance of the subject buildings will be enhanced through the interface with vegetation that cools and filters air being drawn into the buildings. The two townhouses at 14 and 16 Considine Place already demonstrate the intended result expected to be typical for the whole project as they are festooned with vines and other vegetation providing seasonal shade (see photos).

Objectives

In this case study monitoring project at Christie Walk the objective is to determine whether these homes are indeed low energy dwellings, and if so, how much better are they than the SA average.

The project will build on results from two other SENRAC funded projects,

From the baseline study a comparison will be made between the State average and Christie Walk values of,

In addition a comparison is made with 6 monitored homes at the Mawson Lakes concentrating on household peak load that occurs particularly at the time of the SA electricity system summer peak.

3. METHODOLOGY

When the project started 5 homes had been completed. After a few months a sixth home was added. The mix of dwellings include two separate houses and a block of 4, 3 storey Townhouses. Occupancies of the 6 homes are between 1 ­ 2 persons.

Total household energy use for all 6 homes and energy used for water heating was monitored using ETSA Utility installed Email A11 Smart billing Meters. In addition data-logging equipment was installed on two household electricity distribution boards for detailed end-use monitoring purposes. Total Development electricity use will also be undertaken and an average for the site determined. The restriction to two homes for detailed monitoring was aimed at containing data logging costs. Most homes are all electric but some homes have gas cook tops as well, these were not monitored as it was felt that the expense to measure this small end-use was not warranted.

Note: All homes have solar/electric water heaters.

To determine what comfort levels are acceptable to these environmentally aware residents, lounge room and bedroom temperatures were measured by the University of Adelaide in one of the “separate” houses and two apartments. Also a weather station, provided by the University was installed on site to determine the micro-climate however, the data from this was not available at the time of writing the report.

For the comparison at Mawson Lakes there were also 6 homes all of which had detailed end-use monitoring. Only one home was all-electric the rest had gas water heating, cooking and some gas heating. Nevertheless a brief energy, peak load and CO2 comparison for both developments was made. A comparison between Christie Walk and Mawson Lakes is however difficult as the number of people in the 6 Mawson Lakes homes varied between 2 ­ 6 persons, whereas for Christie Walk the variability was 1 ­ 2 person. Where possible adjustments for the larger number of persons and size of homes were made, or at least noted.

4. SIGNIFICANCE OF PROJECT TO SOUTH AUSTRALIA

Residential peak load electricity use in the 5 years to 2000/01 has been growing at about 4.4 % per year. In order to help contain electricity prices and improve residential sector load factor it is important for both new and renovated housing to help contain growth of both base and peak load consumption. The homes at Christie walk should theoretically require little heating and cooling and at the same time maintain good comfort levels within the home. This project determines whether this is in fact the case.

View of three of the Apartments and one of the separate houses

5. ELECTRICITY MONITORING PROGRAM

As mentioned above, electricity use by the homes at Christie Walk was collected in 2 ways;

  1. From ETSA Utilities’ electronic billing meters installed in a group meter box on site. This enabled both peak and off peak (solar water heater ­ SWH) data to be collected from all 6 houses.
  2. From circuit board monitoring at meter boxes attached to 2 individual homes for the provision of end-use data.

In the first case ETSA Utilities provided and installed 1 Email Gateway system that enabled us, using proprietary Email Empwin Software, to down load, whenever required, data in 1ž4 hour intervals. Additionally there are 6, 2 channel, electronic Email A11-L meters that can hold over 3 months of data before overwriting starts to occur.

ETSA Utilities provided us with a training session on how to download data, but used their own equipment in the demonstration. 3 months later when we went to download data again we were unable to do so. After many attempts, the purchasing of different cables and attempting to find a time that ETSA Utilities (EU) could come to us again, it was eventually discovered that EU was using a unique cable that had to be specially made up. By this time several months of data had been lost. Fortunately however, we had the time periods of greatest interest. The winter of 2003, 12/6/03 to 19/09/03 (99 days) and the summer of 2004, 11/01/04 to 29/04/04 (100 days). In addition there was data for April and May 2004 (38 days) which was used to estimate the rest of the year data.

In the second case Measurement Engineering Australia (MEA) supplied and installed end-use data logging on the circuit boards of 2, 3 storey apartments of similar design and area. There was a row of 4 apartments and logging was carried out on the 2 central ones. The logger used was a Datataker DT 600. The power board circuits had the labeling given below. In some cases labeling was quite misleading , for example the circuit marked “fridge” did not have a fridge on it.

House A, (1 person)

House B, (2 people)

Group Meter Box With Email Smart Meters

 

Downloading of data is performed by connecting a cable from box on top RHS to a computer.Note how differently the circuit boards in the two similar houses of similar layout were set up.

For these apartments data was available from 21/6/03 to 31/12/03. After this period mistakes made in downloading and resetting the logger meant data was unfortunately lost for most of the summer period.

Temperature Data

Small portable Hobo data loggers were installed by the University of Adelaide in 3 homes, 2 of which were part of the larger monitoring program. These were the apartment, House A, and the other in separate dwelling House 4. The temperature sensors were placed in the downstairs living area and upstairs bedroom. Data was available for:

Problems Encountered During project

As with most projects unforeseen problems occurred during the study period which had a significant effect on data availability for analysis. The main source of the problem was the inability to find someone connected with UEA that really understood about data monitoring and retrieval. (It was an unrealistic request.) It is strongly recommended for anyone undertaking a data monitoring program to pay extra and get someone that is a professional in the area of data retrieval who is responsible for regularly providing a good and complete data set in usable format. Trying to cost cut in this area does not work it only leads to frustration and data loss!

6. RESULTS

Site Monitoring

Data from the 6 ETSA Utility electronic Email A11-L, 2 channel billing meters was used to determine Summer and Winter electricity use at the site. Since a whole year of data was not available, data from April and May was used to fill in the gaps in order to obtain an estimate of annual electricity use. (There was 9 months of data available and so average electricity use for the remaining 3 months had to be estimated using April/May data. Though this is not ideal it is felt that the error is likely to be small and will not change the overall conclusions of the paper.) Tables 1 to 3 display the results of electricity use per season and year for Peak and Off Peak periods, average daily use over the same time and annual costs and carbon dioxide emissions.

Table 1: Electricity Use (kWh) in Summer, Winter, Aril/May and an estaimate for the whole year for General Power and Off Peak (SHW)
period Summer Winter Apr/May Year  
days 100 99 38 166  
  kWh  
House Code Peak O/P Total Peak O/P Total Peak O/P Total Peak O/P Total No. of
People
1 362 4.3 366.5 484.7 195.0 679.7 155.1 42.0 197.1 1524.3 382.7 1907.1 1
2 639 0.0 638.9 761.7 449.1 1210.8 264.1 58.1 322.1 2554.2 702.8 3257.0 1
A 318 0.0 318.1 422.5 388.9 811.4 108.3 36.9 145.2 1213.9 549.9 1763.8 1
3 325 0.0 324.6 296.2 469.7 765.9 114.2 149.6 263.8 1119.7 1123.3 2243.0 2
B 773 60.9 834.1 905.1 922.4 1827.5 334.4 280.3 614.6 3138.9 2207.7 5346.6 2
4 435 5.9 441.1 n/a n/a n/a 520.9 56.3 577.2 n/a n/a n/a 2
Ave kWh 475 12 487 574 485 1059 250 104 353 1910 993 2903 1.5

Table 2: Electricity use per day (kWh/day) in Summer, Winter, April/May and an estimate for the whole year for General Power and Off Peak (SHW)
period Summer Winter Apr/May Year  
days 100 99 38 166  
  kWh  
House Code Peak O/P Total Peak O/P Total Peak O/P Total Peak O/P Total No. of
People
1 3.62 0.04 3.67 4.90 1.97 6.87 4.08 1.10 5.19 4.18 1.05 5.22 1
2 6.39 0.00 6.39 7.69 4.54 12.23 6.95 1.53 8.48 7.00 1.93 8.92 1
A 3.18 0.00 3.18 4.27 3.93 8.20 2.85 0.97 3.82 3.33 1.51 4.83 2
3 3.25 0.00 3.25 2.99 4.74 7.74 3.01 3.94 6.94 3.07 3.08 6.15 1
B 7.73 0.61 8.34 9.14 9.32 18.46 8.80 7.38 16.17 8.60 6.05 14.65 2
4 4.35 0.06 4.41 n/a n/a n/a 13.71 1.48 15.19 n/a n/a n/a 2
Ave kWh 4.75 0.12 4.87 5.80 4.90 10.70 6.57 2.73 9.30 5.23 2.72 7.95 1.5

Table 3: Average electricity use for the 1 and 2 person homes plus annual cost and Carbon dioxide emissions
period Summer Winter Apr/May Year
days 100 99 38 166
  kWh Cost $ kg CO2
  Peak O/P Total Peak O/P Total Peak O/P Total Peak O/P Total
Average for period
1 person house 411 1 412 491 376 867 160 72 232 1603 690 2293 461.69 2543
2 person house 511 22 533 601 696 1297 323 162 485 2129 1665 3795 616.42 4209
Average per day for period
1 person
house 		4.11 0.01 4.12 4.96 3.79 8.76 4.22 1.89 6.11 4.39 1.89 6.28 2.78 15.32
2 person house 5.11 0.22 5.33 6.07 7.03 13.10 8.50 4.26 12.77 5.83 4.56 10.40 3.71 25.36

 

Table 4: Summary Table of Electricity Use in the 1 & 2 Person Homes at Christie Walk
No. of People Summer (kWh)
(100 days) 		Winter (kWh)
(99 days) 		Year Estimate (kWh)
(using April/May data for remaining days)
  Peak SWH Total Peak SWH Total Peak SWH Total
1 411 1 412 491 376 867 1603 690 2293
2 511 22 533 601 696 1297 2129 1665 3795

It is interesting to note that even though electricity use for water heating is now low, it is still 30 % of total electricity use in the 1 person homes and 44 % in the 2 person homes. This is because all other end-uses have low consumption as a consequence of occupants having energy efficient appliances and the heating and cooling requirement is low. Later in the report it will be discussed how the water-heating component can be further reduced.

Based on electricity Tariffs for Summer and Autumn 2004 an estimate was made of electricity costs. In addition carbon dioxide emissions were calculated based on an emission factor for SA of 1.109 kg CO2/kWh for electricity.

Table 5: Cost per year and Carbon Dioxide Emissions
No. of People Electricity Use (kWh/y) Cost/year ($) CO2 emissions (t/y)
1 2293 462 2.54
2 3795 616 4.21

Though Christie Walk has low occupancy dwellings, this is becoming more of the norm. In SA about 26 % of the population live alone and 35 % in 2 person dwellings, ie 51 % of the population have 1-2 person occupancy homes and only about 15 % are in homes with 4 people (including children). In Germany the situation is even more extreme currently. There, 37 % of the population live alone and 34 % in 2 person homes ­ a total of 71% in 1 or 2 person homes, with only 11 % of homes with 4 people. Therefore for the future what is being developed at Christie Walk is very relevant.

Comparison of Results with other work

A comparison of these results was undertaken with other work carried out previously by the author. In particular a comparison with results from the SENRAC funded report No. 14/00 “South Australian Residential Sector Baseline Study” will be presented. This report used ABS 1997 to1999 survey data to calculate a base line for energy use in SA as a function of number of people in a home and fuel type, gas or electricity. Table 6 shows the results for All Electric homes, such as those at Christie Walk. (It is most likely that average electricity and gas use in 2004 will be higher than that found for 1999, however the base line data is a good reference point.)

Table 6: 1997 - 1999 ABS Baseline Electricity Use in SA for All electric Homes.

No. of People Electricity Use (kWh/y) CO2 emissions, t/y
1 5469 6.07
2 7359 8.16

It can be seen that the homes at Christie Walk performed much better than the State average for all electric homes, with about a 60 % improvement for the 1 person home and 50 % for the 2 person. The improvements in the main are due to the solar water heater, minimal heating and cooling requirement together with installed energy efficient appliances.

There will be significant electricity cost reductions also, but these will not be as large as the electricity reductions due to supply charges remaining the same and a slightly different mix of peak to off peak tariff usage.

Using data from this same report a comparison is also made with 1 and 2 person homes that are All Gas, ie the people use gas for water heating, space heating and cooking. (Gas MJ have been converted to kWh by dividing MJ by 3.6)

The emission factor for gas used was 0.0742 kg C02/MJ and again 1.109 kg CO2/kWh for electricity.

Table 7: 1997 - 1999 ABS Baseline Electricity Use in SA for All gas Homes.

No. of People Electricity Use (kWh/y) Gas Use (kWh/y) Total Energy Use (kWh/y) CO2 emissions (t/y)
1 2609 5523 8133 4.37
2 3676 8321 11997 6.30

It can be seen that the Christie Walk homes performed significantly better than State average all gas homes as well with a reduction of 72% and 68 % in energy in 1 and 2 person homes and 42 %, 33% in CO2 emissions. In addition there will be significant cost reductions.

In June 1999 a report on “Energy Consumption in Small Households” investigated the impact of energy efficient appliances and solar water heaters on energy consumption in a group of low income public housing tenants. The homes, built by the South Australian Housing Trust were constructed along energy efficiency lines and the small (~ 90 m2) 2 storey town houses occupied by 1 person had an electricity use of 2127 kWh/y which is almost identical to that for Christie Walk as shown in Table 5. Indicating that results can be replicated across sections of the community with improved house design and use of efficient appliances. The Public Housing SAHT homes sparingly used small efficient air-conditioners for heating and cooling and were provided with an education program ­ however again, as at Christie Walk, the sample was small.

Load Curves

The quarter hour data provided by the billing meters provided quite a wealth of information about the site. This data was much easier to retrieve and use than the end-use data from the data logger and inherently a cheaper way to go rather than monitoring each circuit of the house. Only where there is a definite need to know accurately the breakdown of appliances in the home is the extra effort worthwhile.

Solar Water Heaters

The billing meters are 2 rate with one rate ­ off peak - dedicated to water heating, and in this case to solar water heating.

It was soon found once analysis started that the solar water heaters were being given a daytime boost in the middle of the day. This is not only bad for optimising solar output from a SWH but it is also bad from a utility point of view, as the customer is paying off peak rates during the peak period and also contributing to peak loads. One of the outcomes of this project is to ask ETSA Utilities to change the electrical boost time clocks.

Additionally it was found that because modern water heating elements are now quite large, in this case ~ 3.6 kW, the tank of water heats up in just a few hours and there is therefore rarely the need for the whole 7 hr boost period. As a result before the booster shuts off round 7 am it restrikes, comes on again, to cover any heat losses. See Figure 1. This effect is common to all monitored water heater projects that the author has been involved with in the past few years and the number of restrikes increase in winter. If the restrike is prevented electronically or the booster switch-on time is pushed forward a few hours, electricity use could be significantly reduced and still the hot water requirement maintained. In addition insulation standards should be further improved. Figure 1 also shows quite clearly the day­time electricity boost. It is noted that during summer only half the residents turned off their electric hot water booster. (See Tables 1 & 2).

Figure 1: Electric Booster Restrikes before 7am and Daytime SWH Boost.

On the SA electricity system peak load day, 14 February 2004, there was no solar water heating boost contribution to peak load from the Christie Walk development.

Peak Power ­ Non Water heating Load

This section looks at an analysis of electricity use on what used to be called the “General Power and Light” tariff of the billing meter. It is now is called T110 or more simply “Peak Power” tariff as opposed to the hot water “Off Peak” power tariff. The results are presented in two parts.

  1. Winter: The winter peak load day is plotted for each home in Figure 2. Because each home had their peak load on a different day no average peak load is given. The average load for the period June ­ September (99 days) is however given for each home and this time for the site as well. (There was no quarter hour billing data available for House 4 during winter.)
  2. Summer: The load for each home is plotted in Figure 3 for, 14 February 2004, which is when the SA electricity system had its peak load. In addition, the average for the 100 day period January ­ April is also presented.

Figure 2: Winter Peak and Average Loads at Christie Walk.

Figure 3: Loads on Summer Peak Day and Average Loads at Christie Walk

Note scale for the Winter peak days is different to the other graphs, this is because the winter peak loads are higher than at other times and it can be seen that most homes are using some supplementary heating which ranges from about 1 ­ 3 kW. However, the average winter load profiles show that heating is not used very often as the average peak load for the 5 homes is, over the whole period, only 0.4 kW.

In summer on the SA System peak day there is only one home that appears to be using any sort of cooling and that is being used overnight and between 14:15 and 15:15 only. At the time of the SA system peak on 14 February the 6 Christie Walk homes are only contributing an average of 0.44 kW per household to peak load. Whereas, see later, it is likely that the 6 homes being monitored at Mawson Lakes were contributing about 4.0 kW/HH, (The contribution is still 4.0 kW/HH if just the 2, 2 person homes are considered). State average on a peak load day for the residential sector is ~ 2.5 kW per home.)

It is interesting to contemplate what the residential contribution on a peak load day would be if all 600 000 residential customers used respectively 0.44 kW/HH, 2.5 kW/HH and 4.0 kW/HH. Their contribution to the peak would be 264 MW, 1500 MW or 2400 MW.

The average peak summer load over the period January to March at Christie Walk is 0.34 kW at 20:30.

Monitoring of Individual Homes

As mentioned previously data for the 2 monitored 3 storey Townhouses of similar construction and layout was only available from 21 June 2003 to 31 December 2003, a period of about 6 months. One home had just one occupant, who is a very committed environmentalist, and consequently energy use in this home is quite low. The second home has 2 occupants that have quite a wide range of appliances and not quite as dedicated a commitment to saving energy, demonstrating that behavior has quite an impact on energy use.

Below are a selection of pie charts that show how electricity use is divided between the individual household circuits for the whole 6-month period and for the individual months of July and December. In addition a load profile for a cold day in June and hot day in December is given.

House A: (One Occupant, Area ~ 90 m2)

Pie charts for the House A electricity circuits are shown below. There were more circuits than usual on the distribution board and each floor, ground (G), 1st and 2nd are separately monitored. There was no monitoring for HW but this was covered by the billing meter data. For this home hot water was ~ 30 % of electricity use.

Electricity use was divided as shown below ­ note percentages are of low consumption, ie in December the smoke detector used 5kWh in the month:
Smoke Detectors 2-5%
Lighting 8-14%
Power Grd Fl (A) 28-39%
Power Grd Fl (B) 16-41%
Power 1st Floor 8-14%
Power 2nd Floor 0-12%

House(A)

Ground Floor and Upstairs Bedroom

Interior House (4)

It can be seen that most of the power use is by the General Power Circuits on the Ground floor ­ which is the kitchen, laundry and living area. Power G(A) has the refrigerator on it and other kitchen appliances and Power G (B) the “entertainment” system and possibly a small heater.

Figure 4: House (A) Cold and Hot Weather Appliance Usage

The graphs showing energy use on a cold winter’s and hot summer day for House (A) are quite interesting. In the winter graph it can be seen that the fridge circuit, Power G(A), has another appliance attached to it. The fridge is always easy to identify because it regularly cycles on and off. It does not appear that any heating was used on this day as loads are too low.

On the hot summer day perhaps no-one was home because virtually no appliances were used and the fridge cycled regularly. However, the graph shows quite dramatically how much standby load was on. In winter the standby was not as great but still quite evident. Note: Scales on the summer and winter graphs are different to highlight standby in summer.

Annual energy use in this home was only about 1763 kWh/y (including SHW) and so standby probably is one of the larger loads ­ at least 15 % of total electricity use. The consumption in this home is about 68 % less than the State average for a 1 person All Electric home.

House (B): (2 Occupants, Area ~ 90 m2)

House (B) had even more circuits on its distribution board than House (A) ­ 12 of them including hot water. For this home SHW was ~ 40 % of electricity use. To be consistent in the pie charts between House (A) and (B) the SHW was left off.

Electricity use was divided as shown below:
Spa 1 - 4%
Lighting 13 - 16%
Oven 1 - 2%
Fridge 2 - 3%
Power Grd Fl 24 - 33%
Power Grd Fl Sunroom 26 - 37%
Power 1st Fl 3 - 7%
Power 2nd Fl 2 - 11%
Power Bench 17 - 28%

The circuit board names did not really relate to what was on the circuits. The fridge was actually on the Power Bench circuit. As before most electricity use was on the Ground Floor in the kitchen, laundry and living areas. The “sunroom “ circuit has a large standby load as has the Ground floor Power. These include a dishwasher and washing machine. There was an intention to identify appliances on each circuit for both homes, however, it was not done in time for this report and will be done later.

Figure 5: House (B) Cold and Hot Weather Appliance Usage

In June, 2 graphs were drawn, one with and the other without solar hot water. (There was no SHW electricity boost in December.) The boost pattern for the SHW was unusual in that not only did it come on during the day but in the evening around 7pm as well, indicating that there may be a manual boost button.

The top graph is repeated with SHW removed so that more detail can be observed on an expanded scale. Most of the load is during the daytime and very late at night. Electricity usage in this home was unusual and perhaps occupants have some shift work. The main loads on the winter day are 2nd Floor lights (grey), Power G (yellow) which could include heating and entertainment plus some standby (see summer graph also) and the oven (purple).

On the hot summer day the load is mostly the fridge and standby with a small load in the middle of the day that could be entertainment plus microwave. Although this home has a spa and dishwasher they are not used very often.

Annual electricity use for House (B) was 5347 kWh/y which is 28 % less than the State average for a 2 person All Electric home. The solar water heater would represent about 15 ­20 % of the 28 %.

Comparison of Performance of Christie Walk Homes and Mawson Lakes Homes on Peak Load Days

The Sustainable Energy Centre at the University of South Australia, Mawson Lakes Campus, has been monitoring the performance of 6 homes for the last few years. Presented below is data for the peak load day in February 2003. The data for 2004 was not available at the time of writing, however, 2003 was slightly milder than 2004 so will not compromise the results. Details of the homes are given on the graphs in Figure 6 but are also tabulated below. It is expected that the peak day results will be typical of many new estates currently being built.

Mawson Lake House Details
House Number No. of Storeys Number of People Area
1 1 3 123 m2
2 1 2 102 m2
3 2 2 182 m2
4 1 3 172 m2
5 1 3 138 m2
6 2 5 217 m2

House 6 is the only All Electric home with a standard storage electric water heater.

Christie Walk House Details
House Number No. of Storeys Number of People Area
1 3 1 90 m2
2 3 2 90 m2
3 3 1 55 m2
4 3 2 120 m2
A 3 1 90 m2
B 3 2 90 m2

It can be seen that the occupancy of the Mawson Lakes homes that were monitored is higher than at Ch*ristie Walk, in fact there is not much similarity between the homes, however, the peak day comparison is still interesting. See Figures 6

Figure 6 : Electricity Usage in 6 Mawson Lakes Homes on Peak Load Day February 2003

All of the Mawson Lakes homes had an air-conditioner, one, House 2, had an evaporative air-conditioner, the rest were RCACs. All air-conditioners were on for some or all of the peak load period from about 12:00 to 18:00, (See Figure 7.)

The evaporative system contributed by far the least load. Most of the air-conditioners were working flat out for most of the time, with main cycling occurring in the evening, though 2 had minor cycling during the day. Their contribution to peak load varied from ~1kW (evaporative) to ~ 8 kW for House 6 (RCAC) with an average of ~ 5 kW. At the time of the peak load ~ 13:00 the average contributionby the 6 Mawson Lakes homes was ~ 4.0 kW/HH.

It is interesting that only one home, House 5, did not have an RCAC on during non working hours. A look at what appliances were on during the day shows that little, besides the refrigerator and standby were on together with the air-conditioner. In fact the non air-conditioning load was not much more than at Christie Walk, even for the 5 person HH.

It is known that energy efficient appliances have been installed at Mawson Lakes and so if the homes were better designed for cooling they would have quite a good energy performance. However, home owners in Mawson Lakes use quite a lot of gas as well as electricity.

In the University of SA Report, “Development, Implementation and Promotion of a Score Sheet for Household Greenhouse Gas Reduction in South Australia”, that looks at the average gas and electricity use in a few hundred homes, it is found that the average energy use in the sub division (average HH size 2.7 and average area 172 m2) is:
Electricity 5853 kWh/y
Gas Use 8194 kWh/y
Total 14047 kWh/y
CO2 emissions 8.68 t/y

Table 8 below summarises the results for Christie Walk and Mawson Lakes.

 
Location Ave No
People 		Electricity
kWh 		Gas
kWh 		Total
kWh 		t CO2/v Area
m2 		kWh/capita
per y 		t CO2/
Capita
per y		 kWh/
Capita
per y*m2		 t CO2/
Capita
per y*m2
CW 1.5 2903   2903 3.22 90 1935 2.15 21.50 0.0239
ML 2.7 8194 5893 14047 8.68 172 5203 3.21 30.25 0.0187
CW/ML 44.4%         47.7% 62.8% 33.2% 28.9% -27.6%

The Table indicates that the homes at Christie Walk (CW) have:

The reason for the carbon dioxide emissions/capita/m2/y being more is that the emissions are being divided by 2 large numbers. Analysis in the Base Line Study Report for SA showed that the parameters that significantly impact on household energy use are, number of people in the home, appliance mix and probably behavior. Though area of house was significant in determining energy use the significance did not change markedly as the number of the rooms in the home increased from 5 ­ 9. Similarly, though income was significant the significance on energy use did not change for people in the first 8 decile ranges, ie for people in the lowest 80% range of incomes. The above shows that for SA a better indicator of how a house performs is a per capita measurement rather than a per capita per unit area measurement - which is not what is usually done. More work needs to be undertaken in this area.

Energy operating costs at Christie Walk would also be less than at Mawson Lakes, but no estimate was made of by how much.

(Two additional notes to make about the Figure 6 graphs is that the anomaly circuit of the 6 homes is that marked "heater" in House 3, which is unlikely to be a heater on a day of extreme heat. It could be supplementary portable cooling or ceiling fans.

The electric water heater in House 6 undergoes a number of restrikes to account for heat losses even though the weather was hot. It may mean that the thermostat is set very high. )

To complete this section Figure 7 plots the NEMMCO SA electricity load curve for the peak day in February 2004. Note that the peak extends from about 12:00 to 18:00. Superimposed on this curve is a load curve for the average of the Christie Walk homes on the same day. Added also are the average of ~ 50 homes at Mawson Lakes on a peak load day in 2001, 2004 data was not available at time of writing. However, it is unlikely that there will be much significant difference between the years. The graph highlights the relative contributions to peak loads that the 2 developments make.

Figure 7: SA Electricity System Peak Load Day Feb 2004 and Average Electricity use at Christie Walk on same day. (Mawson Lakes data is for a previous year.)

Room Temperature

Dr Veronica Soebarto from the University of Adelaide installed small Hobo temperature and humidity data loggers in 3 homes. They were located in the living area and main upstairs bedroom. 2 of the homes were part of the energy study group, the apartment labeled House (A) and the separate dwelling labeled House 4. The 3 rd dwelling became occupied some time after the others and was not included in the project.

For house (A) the temperature data covered the period, May to September 2003 and December 2003 to March 2004. For House 4 bed room and living area data was available from May to October 2003 and temperature of the main bedroom, only, from October 2003 to March 2004.

Results are presented in Figures 8 & 9 and a summary given in the table below.
Sensor Location Upstairs BedrmTemperature
(Deg C) 		Living Room Temperature
(Deg C)
House A    
Summer 20 C- 35 C (2 days >32C) 19 C ­ 32 C (1 day 32C, mostly < 28C)
Winter 14 C* ­ 22 C 14 C* ­ 20 C
House 4    
Summer 20 C ­ 36 C (5 days >32 C) N/A (expect ~2 days >32C)
Winter 16 C* ­ 22 C 16C* ­ 20 C

*Low Temperatures occur mainly in the early hours of the morning

Figure 8: House (A) Bedroom and Living Room Temperatures

As mentioned previously House (A) is a 3 storey Townhouse in the centre of a 4 Townhouse row. The living area stayed fairly comfortable in this home without cooling during the almost 2 week hot spell in February when temperatures reached 42 C. Mostly the living area temperature was below 28 C though it reached 32 C when outside temperatures were 40C and 42 C. (See Bureau of Meteorology temperature graph in Figure 9 for ambient temperatures in February 2004). The upstairs bedroom was about 2 ­3 C warmer but would have only been unpleasant on 2 days, when a ceiling fan could have been used for sleeping.

In winter the living area, when occupied, was mainly between 16 ­ 20 C, a bit cool at times but comfortable for the occupant who is happy to put on extra jumpers rather than heating.

Figure 9: House (4) Bedroom and Living Room Temperatures and BOM Adelaide Airport Temperatures for February 2004

In the separate House 4 there were only upstairs bedroom temperatures available for summer. This room is likely to be 2 ­ 3 C higher in temperature than the living area. Bearing this in mind it was marginally hotter than the row House (A) and probably over 32 C on ~ 2 days.

In general House 4 is slightly warmer in both summer and winter than House (A) with living room temperatures between16 ­ 20 C but with a slightly higher average and lower temperature fluctuations. Again the occupants are prepared to put on extra clothing rather than heating.

In both homes the chimney venting effect of the small stairwell helps to remove hot air from the downstairs rooms. For the couple of days a year that perhaps supplementary cooling could be nice, personal fans are adequate and considerable cheaper to purchase and run than evaporative or refrigerative air-conditioners.

Unfortunately there were no temperature measurements in homes at Mawson Lakes. Perhaps this should be done later.

CONCLUSIONS AND RECOMMENDATIONS

There is no doubt that the homes monitored at Christie Walk are low energy homes. They performed significantly better than the State average and when compared with the homes at Mawson Lakes, especially if energy use and CO2 emissions are calculated on a per capita basis. However, they do not rate as well when emissions are calculated per unit area as well as per capita. This is because the average house size at Mawson Lakes is almost double the average at Christie Walk, and so dividing by a large number gives an apparent advantage to the larger homes. Separate work by the author indicates that up to a certain house area ­ value unknown but up to a 9 room house ­ area is possibly not a significant determinant of energy usage. It is an topic where more work is needed.

Monitoring of the Solar Water Heaters resulted in realising that they are being a day time electricity boost on the off-peak tariff and it is recommended that this should be rectified. In addition the solar water heaters are restriking during the overnight boost period indicating that losses are still significant even after MEPS (Minimum Energy Performance Standards) improvements. Therefore the recommendation is that standards for allowed water heater losses should be further strengthened. In addition the boost period is too long and electricity consumption by water heaters could be reduced probably by up to 30 % by moving the boost time to earlier in the morning and preventing by electronic means, boost restrikes.

Though it was the intention to undertake an appliance survey of homes with detailed end-use monitoring, this was not done for lack of time. Nevertheless, it is believed that not much extra useful information would have come out of such a survey as neither home had large contributions to peak load. The most interesting thing to come out of the end-use monitoring was the large contribution that standby loads make to low energy, and for that matter, any home. End-use monitoring is very intrusive on people’s lifestyles and privacy and it is not a good thing to do too often!

As in most homes in SA energy use at Christie Walk was greater in winter than summer. Though there was virtually no summer cooling on the site there was a little winter heating, but much less than the norm, also higher loads occurred in winter than summer. The summer contribution to the electricity system peak day was very low, ~ 0.44 kW compared to the State average of 2.5 kW and the Mawson Lake average of 6 homes of 4.0 kW.

Between 4 ­ 18 February 2004 there were virtually 12 days where the temperature was 32 C or higher and there were 2 days over 40 C. Temperature studies of 2 homes showed that in that time the living area of House (A) was over 32 C only once ­ on the 42 C day - and was mostly 28 C or less. The house does not have any cooling. Upstairs it was about 2 degrees hotter. The upstairs temperature of House (4) was marginally hotter than that of House (A) and unfortunately no living room temperatures were available for this home during the hot spell. The University of Adelaide is undertaking more detailed temperature studies at Christie Walk and their results will become available publicly. A comparison with temperatures in Mawson Lakes homes would be interesting.

Note: Mawson Lakes was used as a comparison site as data was available there. It is expected that analysis from other new housing developments would produce similar results.

The intention was to eventually make some of the data from the Christie Walk development available for web based educational purposes. It is still hoped to do this at a later stage.

Finally it is noted that if all 6 houses at Christie Walk would have installed 1 kW of PV on their roofs then on the peak load day, the 0.4 kW contribution by the homes to peak load would have been reduced to zero and a small amount of electricity would have been exported to the grid. Additionally, annual electricity use and CO2 emissions would have reduced a further 15% ­ 20 %.

REFERENCES

Saman, Wasim A/Prof, Mudge, Lachlan June 2003 “Development, Implementation and Promotion of a Scoresheet for Household Greenhouse Gas reduction in South Australia”. Sustainable Energy Centre University of SA. Report to Elizabeth Young, Partnerships Group, AGO

Saman, Wasim A/Prof, Mudge, Lachlan, 1999 “A Program for Evaluating and Monitoring the Energy Consumption in a Major Suburban Housing Development”, SENRAC Report 9/99 Sustainable Energy Centre University of SA.

Oliphant, M. 2003, South Australian Residential Sector Baseline Study Of Energy Consumption: Final Report, Sustainable Energy Centre, University of South Australia, Mawson Lakes, Adelaide. SENRAC Report 14/00

Oliphant, M. 1999 “ Energy Consumption in Small Households”. Report for the Australian Electricity Supply Industry Research Board.

9. ACKNOWLEDGEMENTS

Acknowledgement is given to the other participants in this project:

Dr Paul Downton (UEA) and all the Residents from Christie Walk who allowed their homes to be monitored and data on their electricity consumption to be collected and analyzed. All the analysis and text for this report was been done by the author except for the introductory description of the Christie Walk project and the front page drawing that was done by Paul Downton. He also provided the photos, which I am very grateful for.

Brian Forte (IT Support): It was good to have someone stay around long enough to collect some usable data. His help was valued.

Dr Veronica Soebarto (University of Adelaide)

The use of the temperature data from the Hobo meters placed in homes were very helpful to the report.

A number of problems occurred initially to set this SENRAC project behind schedule. Also limited budget and resources made it difficult to do as in depth a study as I would have liked. In other projects I have found that when people move into a new house it takes at least a year to settle in and come to a stable energy use regime. We did not have the luxury of waiting in this project. Nevertheless I believe that the results are sufficiently interesting to warrant a more detailed follow up study later. The funding from SENRAC was appreciated.

Though, realistically, the use of mud brick and straw bale homes are unlikely to become the norm in SA the project will show whether well insulated and orientated homes are effective in reducing thermal demand and also increase the penetration of these low impact on the environment homes.

26 October 2006