Passive house vs. passive design: sociotechnical issues in a practice-based design research project for a low-energy house

ABSTRACT Building performance simulation tools such as the Passive House Planning Package (PHPP) can be invaluable for improving energy-efficiency in housing design. However, achieving improved energy performance is also a sociotechnical issue, and how this is dealt with during the architectural design process seems less well studied. This collaborative design research project for a low-energy prefab house with an industry partner, a manufacturer of Structural Insulated Panels (SIP), is used as a case study to show that it is possible to achieve high energy performance while addressing specific socio-technical concerns within an Australian volume homebuilding market. A key issue that emerged in this project was the perceived tension between passive design expectations in Australia and those promoted through the Passive House software tool.


Introduction
Building performance simulation tools can be invaluable during the design process to improve the energy-efficiency of proposed buildings. The available body of literature on energy performance of buildings is extensive, exploring both technical aspects such as the efficiency of building envelopes (De Boeck et al. 2015), as well as social aspects such as user evaluations and preferences (Hauge, Thomsen, and Berker 2011). How sociotechnical factors play a role in the design process itself, however, seems less well studied and understood. In order to effectively integrate performance assessment tools into architectural practice, it would be important to examine how these factors interact during the design process, beyond a focus on energy-efficiency alone, which has been criticized for example by Shove (2017).
The focus of this paper are sociotechnical issues that emerged during a collaborative and industry-linked, government-funded design research project (LP13). The LP13 project was set up as a collaboration between an industry partner, a manufacturer and supplier of Structural Insulated Panel (SIP) construction systems, and the Innovation in Applied Design Lab (IAD Lab) at the University of Sydney with an integrated multidisciplinary team of researchers in architecture and engineering. 1 The aim of the project was to develop an energy-efficient and cost-effective prototype dwelling for the Australian volume housing market, to be built with the industry partner's wall system. Such a prototype could help demonstrate the effectiveness of a prefabricated SIP walling product as part of an overall more energy-efficient offering. After the construction of a smaller prototype of the SIP walling system in the factory (Figure 1), the project ultimately did not proceed to construction of the actual building because of changes in the commercial circumstances of the industry partner. However, the design development described here still CONTACT David Kroll krolldavid@hotmail.com; david.kroll@unisa.edu.au 18a Beatrice Street, Prospect SA 5082, Australia provides a valuable case study of socio-technical issues that arose during that process. In practice, architectural design balances multiple factors and interests such as site, orientation, flexibility (for future use), customer expectations, planning guidelines, development covenants, building performance, as well as technical and construction requirements. If energy-use minimization is approached from a purely technical point of view based on performance simulation, then the most efficient form would be as compact as possible. However, as this case study illustrates, the problem can be more complex in practice.
The LP13 project was influenced by complex demands of the volume house building industry and specific requirements from the industry partner. The interdisciplinary team inevitably involved negotiations between disciplinary ambitions and ideals. Tensions between technically desirable solutions and user requirements were part of this process. As the industry partner's client and target market, the volume house builder and its customers as end-users also indirectly played an important role. A significant sociotechnical dilemma that emerged during the design development in the LP13 project were discrepancies between the ideal solution suggested by the PHPP software tool of a compact form and isothermal environment on the one hand, and the local thermal habits and preferences of an 'indoor/outdoor' lifestyle in an Australian market on the other. The design outcome of the project was a hybrid approach that combined benefits of both passive housing and passive design.
In terms of methodology, the study takes a practice-based research approach. This approach draws on the reflective practice model (Schön 1991) by examining knowledge that emerges from real-world experience during the design process. The practice-based approach meant that the authors were themselves part of the design process rather than outside observers, as is often the case in traditional scientific research (Cross 1993). This 'insider' perspective could be considered a limitation of practice-based research but, in this case, the reflection on real-world dynamics can serve to highlight issues within complex socio-technical environments that may be overlooked in other research. In this case, the socio-technical tensions between the energy-efficiency assessment tools and members of the project team is a problem that emerged during the design development and that needed to be addressed in the proposed design solutions.
The project was developed through bi-weekly co-design workshops where progress was discussed and solutions were developed. Each discipline focussed on part of the overall problem and developed improved solutions within the boundaries set by the house builder and market. The SIP manufacturer led the development of the prefabricated wall system. The architectural designers researched the target market, the site design guidelines and restrictions, and developed design solutions in collaboration with the other team members. The engineering team tested the performance of the SIP, provided feedback and, together with the architectural team, simulated the overall building performance.
The Passive House standard was identified as one of the criteria to verify the aim of achieving a high level of energy performance. The Passive House Planning Package (PHPP) was used to assess the energy performance of the building and to ensure that the design would be able to meet the criteria of a stringent performance standard. The number of completed buildings to Passive House standard has grown in recent years in Australia. The Passive House database currently counts 19 certified buildings across Australia, many only constructed in the last few years (Passivhaus Institut 2019). Yet, the approach still encounters a level of scepticism in Australia about its suitability for local climates, evidenced, for example, in the media in articles about the Passive House standard (Marlow 2016). Further research would be beneficial to better understand what caveats and adaptations to the Passive House approach may be useful for Australian contexts.

Passive house vs. passive design
Previous studies indicate that occupant behaviour in warmer Australian climates can be at odds with approaches to energyefficiency that aim to create an isothermal environment for the whole house (such as Passive House). A study from 1991 shows that very few people in Adelaide heated (only 4%) or cooled (only 11%) the whole house but that the vast majority relied on conditioning selected areas like the bedroom or living room only when needed (Coldicutt, Williamson, and Penny 1991, 257-258). While these choices may simply reflect existing habits and do not necessarily mean they are appropriate in buildings with highly efficient thermal envelopes, such learned behaviour is still a social factor which seems likely to have an influence on user preferences and understanding of energy efficiency in dwellings in Australia. Another factor is the perceived connection between indoors and outdoors and a desire to be able to open windows for good cross ventilation, which is seen as important by occupants. A study of occupant behaviour of a housing development near Adelaide showed, for example, that ( . . . ) resorting to air-conditioners was the least preferred strategy due to implications for their energy bills. Turning on ceiling fans, opening or closing windows and doors, and opening or closing curtains were the first set of actions taken by most occupants when they wanted to be cooler. (Soebarto and Bennetts 2014, 19) The study found that opening windows was more commonly used as a thermal control strategy than air conditioning (Soebarto and Bennetts 2014, 19).
Such behavioural patterns do not always neatly align with energy performance models and can have an impact on preferred built form. A compact form is less important if the whole house is not usually heated or cooled, and if the aim is to achieve good cross-ventilation and a good connection between indoors and outdoors. Social practice theory recognizes strong links between habits and preferences, and that it is intrinsically difficult to change these habits for sustainability (Shove 2010(Shove , 1276. To make energy-efficient buildings more appealing to an Australian audience, their design would also need to take account of local dwelling ideals of an indoor-outdoor lifestyle. In contrast to energy-efficient design that aims to achieve constant indoor conditions, the above described behavioural patterns of opening and closing windows are often associated with an adaptive approach as '( . . . ) "passive buildings", where the control of temperature is achieved largely by thoughtful climatological design, and by giving control of the thermal environment back to the occupants' (Humphreys, Nicol, and Roaf 2015, 7). Such an approach to sustainable design has a longstanding tradition in Australia and is exemplified by the work of Glen Murcutt, for example. Rather than separating the indoor and outside climate as effectively as possible to minimize energy consumption, the kind of 'passive design' or 'passive building' approach taken by Murcutt aims to create buildings that are connected to the outside, with good natural cross-ventilation, good shading and ceiling fans (Lecaro et al. 2017). His designs promote 'permeability' (Vaughan and Ostwald 2014), often with living areas that are partly outdoors. One of the best-known Murcutt designs, the Marie-Short House (Kempsey, New South Wales, Australia) for example, emphasizes this 'link between the inside and outside' (Lecaro et al. 2017, 2) and integrates two large shaded verandas that provide expanded living space linked to the indoor living and dining rooms.
These local predilections also influenced the design development for the LP13 house. During the design process, the most compact form was rejected in favour of a less compact floorplan that allows for more 'permeability', cross-ventilation and interaction with the outside. These dwelling habits could be a reasons why energy-performance standards that favour constant indoor conditions (such as Passive House) are sometimes seen as conflicting with Australian 'indoor-outdoor' dwelling ideals exemplified in Glenn Murcutt's house designs, where the whole house is not seen as one thermal envelope but instead as separate areas to be heated or cooled when needed. In the design development of the prototype house, such concerns were taken into account. Rather than simply applying the Passive House standard, the design was developed to ensure that both the adaptive approach as well as the constant indoor environment can be accommodated.

Suburban volume housing
The industry partner collaborated closely with a major house builder who operates primarily in the suburban Australian volume housing market (Figure 2). The preferences and demands of the suburban volume housing market therefore played an important role for the LP13 project. The industry partner liaised closely with the volume house builder to develop the project brief for a single-storey house on a suburban site. This target market was a significant part of the social and cultural context which the project depended on.
The environmental impact of low-density detached suburban housing development has long been a cause for concern. From an urban design and planning point of view, this type of development has been criticized for being inherently car-dependent and less walkable than more compact housing (Burton, Jenks, and Williams 2003). More compact medium-density typologies, such as terraced housing, are often seen as a more sustainable and energy-efficient alternative (Moore, Clune, and Morrissey 2013). Medium-density housing has also recently been promoted by planning bodies in Australia through a series of competitions and new guidelines ('Winners Announced: NSW's Missing Middle Design Competition' 2017). From a building envelope point of view, a more compact form is also seen as more efficient and tends to achieve a better energy-performance rating with tools such as Firstrate5 (NatHERS) or PHPP (Newton, Tucker, and Ambrose 2000). The correlation between built form and energy efficiency, however, is also complex and controversial, with some studies suggesting that the compact city idea is too deterministic (Neuman 2005), and that, in certain scenarios, lowdensity housing can be as, or more energy efficient (Ahmadian et al. 2018).
Irrespective of such debates about sustainable densities, the reality of the house building industry in Australia is that the detached suburban house is still among the most common types  of new housing in Australia and often one of the most affordable options for larger or growing households, such as young families (Rosewall and Shoory 2017, 1). The reasons for it may be both cultural and historical, with a long-standing tradition in Australia of comparatively large building plots for its houses (Dalton et al. 2013). This reality of the market was a determining factor for the LP13 project brief in terms of site and typology.

Design of prototype house
The main project task was to develop a low-energy prototype house that would be suitable for the industry partner's target market. Design proposals of a single-storey dwelling were developed for a building plot of 30 × 12.5 m provided by the industry partner based on the house builder's input. The criteria that drove the site and type selection determined by the market that the house builder operated in. The house builder constructed a large number of suburban houses particularly of that typology (single storey house on a 10-15 m wide site) in outer urban suburbs. This site and its dimensions (12.5 × 30 m) were therefore seen as common and 'typical' enough that the design could be potentially be adapted to other sites. The first design iterations for discussion in the team focussed on creating a compact form to optimize energy-efficiency, which would make it easier to achieve a high NatHERS rating and potential Passive House certification (Figure 3).
However, purely technical arguments based on the ideal compact form for energy simulation tools were rejected based on qualitative, more subjective dwelling preferences, related to those mentioned above. During design development discussions, the industry partner and members of the project team raised concerns about initial proposals with a compact form. Less compact proposals with strong links between inside and outside were favoured by most members of the project team as a more acceptable and appealing option for warmer climates in an Australian market. Other factors were the specific qualities of the SIP system. The system achieves higher insulation values at a lower cost per m 2 than traditional construction, allowing for longer perimeters within a similar budget.
To take account of these concerns, options were investigated to understand if a less compact form could offer other advantages. This idea was explored in a series of diagrams, drawing on research about flexible housing (Schneider and Till 2007). The starting point was a compact form with key spaces (e.g. living/dining area, bedrooms) grouped into distinct volumes which could then be pulled apart (Figure 4). Such an arrangement would have several advantages over a simpler and more compact form.
The first advantage identified is 'slack space', seemingly extraneous areas that could be used for example as outdoor living space, for possible future extensions, as a yard or for drying laundry. The second advantage is that the functions within these volumes could be interchangeable if they are sufficiently large. A 30 m 2 + volume could accommodate a living/dining/kitchen area, for example, or 2 × bedrooms + 1 × bathroom, a small studio apartment, a live-work studio or flexible garage. That means the arrangement could support future floor plan changes and be adapted to the best sun orientation. As seen in the diagrams, the volumes would have non-loadbearing internal walls so that the layout could be modified in the future if needed (Ramirez-Lovering 2013). Such a floor plan would challenge the kind of mono-functional suburban housing designed for 'typical' families, but not for most other household types. Each of the volumes could be heated or cooled separately to suit the thermal preferences mentioned earlier, as the walls are constructed with the industry partner's insulated SIP system.
The volumes could be adapted to suit various site dimensions (e.g. of 10, 13 or 15 m width), without invalidating the basic principles. This means that the design could be transferable to other sites within a reasonable range. The first designs based on these diagrams tried to avoid an internal garage and instead provided an option for car parking in the front, for example as a car port. However, many developments require the provision of at least one internal garage in their design guidelines.  A site in Claymore near Sydney in a new suburban development was identified as suitable for LP13. In the Sydney context, this development is at the more affordable end of the market. The design guidelines for the development include requirements for an internal garage, as well as specific setbacks from the site boundaries and street. The initial floor plans were developed and refined to comply with these development design guidelines as well as the Platinum Level of the Livable Housing Design guidelines (Livable Housing Australia 2017).
The prototype floor plan was designed as a family house for 2 adults and 2 children. The house has 3 bedrooms, a living/dining/kitchen area, laundry, bathroom and the required garage ( Figure 5). However, if circumstances change, the house could be adapted to suit other tenancy types. In the future, for example, personal car ownership may not be needed anymore, or one car parking space in front of the house might be considered adequate. In that case, the garage could be adapted to provide an extra bedroom for example. The slack space could also be used to extend the house and to add a bedroom if needed. Another floor plan option is a separate studio apartment in the front. The studio apartment could be used by grown-up children or could be rented to a student or young  couple. Although the design aligns with the prescriptive guidelines of this suburban development, this design would support more diverse suburban densities and tenancy types in the future. The layout of the house would also be age-friendly and its flexibility would make future reconfigurations for downsizing and ageing in place easier.

User customization options
To suit a suburban volume housing market, the prototype was designed to be customizable to personal tastes and budgets. Volume house builders typically offer a floor plan type with several façade and style options. Example variations of the design were developed with different styles for the external  envelope, for example with a hipped roof, a skillion roof or a gable roof -both with garage or without ( Figure 6).
The type that was chosen for Passive House testing was one with a skillion roof (Figure 10). The principles would work with the other types but the skillion roof was preferred for several reasons. One was the possibility of using the slope of a raked ceiling to support night purge ventilation. The other was that a skillion roof type is a common and cost-effective construction type in volume housing in Australia. It was important to prove the concept for a house type that would not require too many non-standard and potentially cost-prohibitive details.

Construction system
The volume house building industry in Australia is traditionally risk-averse and tends to prefer established methods like brick veneer and timber-frame construction (Dalton et al. 2013, 14). Cost of construction per m 2 is a crucial consideration in an industry in which house value is primarily determined by location and size (Clune, Morrissey, and Moore 2012). The industry partner currently offers SIP wall systems with an Expanded polystyrene (EPS) insulation core and Magnesium Oxide (MgO) boards externally. The system uses different 'spline' options to join the panels. One system uses fibre-reinforced plastic (FRP) channels Figure 10. PHPP assessment results with basic configurations failed to achieve the Passive House standard. Figure 11. Energy balance heating for initial configuration that did not meet the Passive House standard. (Figure 7), another uses SIP splines or timber studs to join the panels (Figure 1).
An advantage of the SIP wall system with timber studs as splines is that it is easier to achieve compliance with the Australian building code without the need for more custom structural engineering. The additional structural design costs could otherwise be a barrier and disadvantage over more established wall systems such as brick veneer. For the scope of this case study, the SIP wall system with timber splines was therefore used as the baseline for PHPP testing. Thermal bridging from the timber supports was taken into account. However, further testing would be recommended to assess interstitial condensation risks, which was also expressed to the industry partner.
The tested design uses detailing that can be constructed efficiently with the industry partner's SIP wall system (Figure 8). The skillion roof has insulation between the rafters and below the roof, which was chosen as a cost-effective construction with an overall high insulation value. Thermal bridging could be reduced further if needed.

PHPP software tool
Passive House was chosen as a well-established international standard to test and verify the thermal performance of the LP13 design. The challenge that emerged out of the collaborative design process was to apply the Passive House standard to a less compact house typology within customer expectations of the  volume housing market, which allows for good cross ventilation and connection to outdoors.
In any Passive House building the occupants are expected to understand the way the building should be operated, for example by opening windows for natural ventilation. To minimize reliance on ideal occupant behaviour, however, the house was designed to automatically facilitate natural ventilation through high-level windows with actuators, located below the roof in the corridor and living room (Figure 9). These high-level windows offer secure means of maintaining the ventilation requirements for Passive House certification without the need for occupants to manually open and close windows. Simplifying the building operation in this way should make it easier to pass instructions on to future occupants.
Taking into account thermal bridging from the timber framing, the external envelope with the detailing shown in Figure 8 achieved the following overall U-values: • External wall: 0.341 W/(m 2 K) • Roof: 0.201 W/(m 2 K) • Floor: 3.074 W/(m 2 K) • Windows: 1.70 W/(m 2 K) All thermal conductivities were taken from the PHPP handbook apart from the values for MGO Board, Insulation Batts Bradford Gold HI-P and Insulation Blanket Bradford Anticon 60, which were taken from the manufacturers. This first configuration, however, did not achieve Passive House certification in the PHPP ( Figure 10). The Heating Demand is 41 kWh(m 2 a) and the Primary Energy 176 kWh(m 2 a), clearly exceeding the Passive House criteria. Heat losses through the floor are particularly high, which can be seen in the heat balance graph in Figure 11. 2 After testing different configurations, the following modifications were made to achieve Passive House certification: • 100 mm of insulation added to the floor to achieve U-value of 0.5 W/(m 2 K) • Wall insulation to be improved to achieve U-value of 0.25 W/(m 2 K). • Solar thermal 3 and PV 4 added to the north-facing roof To achieve the minimal heat loss in winter, the added insulation has the effect of increasing the overheating risk in summer. The ventilation and shading strategy therefore involves opening the corridor windows to reduce the risk of overheating. With this ventilation strategy, the frequency of overheating ( > 25°C) is below 10%. 5 With the revised configuration, the Heating Demand could be reduced to 17 kWh(m 2 a) and Primary Energy to 119 kWh(m 2 a) (Figures 12 and 13).

Discussion and conclusion
This industry-linked project illustrates how the challenge of achieving more energy-efficient architectural design and construction is part of a complex socio-technical environment with diverse influences. The project team needed to consider factors such as the context of the suburban volume housing market, user concerns and perceptions, technical properties of the SIP system, results from the PHPP to assess energy performance, as well as site, planning guidelines, orientation, and flexibility of future use. Conflicts between these factors had to be addressed in the design process and technical solutions had to be reconciled with perceptions and behaviour. The LP13 design demonstrates 'trade-offs' that take account of complex sociotechnical concerns raised in the design process and still meet the Passive house criteria.
The design provides an example of how the Passive House standard can be applied to a warm temperate climate on a site near Sydney and combined with the ideal of 'passive design' and 'indoor-outdoor' living. On the one hand, the design incorporates the Passive House concept of creating a consistent thermal environment for the whole house and of separating the indoor and outdoor climate as effectively as possible. On the other hand, the proposal also had to take account of preferences that favoured a less compact form and a more adaptive approach to passive design, with a building envelope that supports a good connection between inside and outside, good cross or purge ventilation, and good shading with outdoor living. spaces The case study also provides an example of how the Passive House standard could be applied to the suburban volume housing market in Australia with a cost-conscious house design using a prefabricated SIP wall construction for a site near Sydney. While the project has to take account of the conditions and demands of a low-density volume housing market, the proposed house design challenges current suburban volume housing typologies with more flexible and adaptable floor plans. The industry-linked and multi-disciplinary process that has been employed here could serve as a helpful example for future low-energy housebuilding projects with opportunities for architects, engineers and other disciplines for collaboration.
While this example demonstrates that the Passive House standard could be achieved even for less compact suburban house types, the project also suggests that performance assessment tools such as PHPP could do more to take account of local user preferences and habits. Adjustments could be enabled in the software tool to account for different thermal preferences and cultural or local specificities. In future research, a comparison between PHPP and Australian specific energy-rating tools like AccuRate would be useful to understand the differences in approach and energy performance.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by Australian Research Council linkage project grant 2013 (LP13).