Scholarly article on topic 'Continuous Mechanical Ventilation in Housing ⬜ Understanding the Gap between Intended and Actual Performance and Use'

Continuous Mechanical Ventilation in Housing ⬜ Understanding the Gap between Intended and Actual Performance and Use Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
Keywords
{"Home use practices" / "mechanical ventilation" / "building performance evaluation" / "user interaction" / "low energy homes."}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Magdalena Baborska-Narozny, Fionn Stevenson

Abstract Improved air tightness and mechanical ventilation systems are regarded as vital elements of low energy strategy in housing. Mechanical ventilation (MV) has become part of the heating energy load optimization model due to its capacity to secure recommended air change levels without depending on daily active control by inhabitants or uncontrolled air leaks. Numerous in- use issues related to continuous mechanical ventilation systems have been identified through field studies. They relate to underperformance of the as built system compared to design targets as well as unintended operation modes. The gap between design intention and actual performance and use of continuous MV in housing context is generally regarded as a threat to expected energy savings or inhabitant's health and as such it should be narrowed as far as possible. Reducing the gap partly depends on the improvement of the continuous MV model to allow for better match with inhabitants needs. This paper proposes a dynamic framework linking factors that influence the emergence of this gap. The framework is based upon findings of previous studies as well as results of a one year-long in-depth Building Performance Evaluation of 40 households in two UK developments. Organizing the pitfalls of embedding MV design into practice in a sequence as well as indicating important links within the process can help to make the complexity more comprehensive and possible to tackle efficiently. Importance of natural ventilation design to backup for MV to increase redundancy and resilience is also highlighted.

Academic research paper on topic "Continuous Mechanical Ventilation in Housing ⬜ Understanding the Gap between Intended and Actual Performance and Use"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Energy Procedia 83 (2015) 167 - 176

7th International Conference on Sustainability in Energy and Buildings

Continuous mechanical ventilation in housing - understanding the gap between intended and actual performance and use.

Magdalena Baborska-Naroznya,b *, Fionn Stevensona

aSheffield School of Architecture, Sheffield University, Sheffield S10 2TN, UK bFaculty of Architecture, Wroclaw University of Technology, 50-370 Wroclaw, Poland

Abstract

Improved air tightness and mechanical ventilation systems are regarded as vital elements of low energy strategy in housing. Mechanical ventilation (MV) has become part of the heating energy load optimization model due to its capacity to secure recommended air change levels without depending on daily active control by inhabitants or uncontrolled air leaks. Numerous in-use issues related to continuous mechanical ventilation systems have been identified through field studies. They relate to underperformance of the as built system compared to design targets as well as unintended operation modes. The gap between design intention and actual performance and use of continuous MV in housing context is generally regarded as a threat to expected energy savings or inhabitant's health and as such it should be narrowed as far as possible. Reducing the gap partly depends on the improvement of the continuous MV model to allow for better match with inhabitants needs. This paper proposes a dynamic framework linking factors that influence the emergence of this gap. The framework is based upon findings of previous studies as well as results of a one year-long in-depth Building Performance Evaluation of 40 households in two UK developments. Organizing the pitfalls of embedding MV design into practice in a sequence as well as indicating important links within the process can help to make the complexity more comprehensive and possible to tackle efficiently. Importance of natural ventilation design to backup for MV to increase redundancy and resilience is also highlighted. © 2015 The Authors.Publishedby Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of KES International

Keywords: Home use practices; mechanical ventilation; building performance evaluation; user interaction; low energy homes.

* Corresponding author. Tel.: +44 114 222 0319; fax: + 44 114 2220315. E-mail address: magdalena.baborska-narozny@pwr.edu.pl

1876-6102 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of KES International

doi:10.1016/j.egypro.2015.12.207

1. Introduction

Highly energy efficient building models incorporate increased air tightness of buildings as a part of a 'fabric first' approach to limit heating demand. Improved insulation comes first followed by reducing heat losses through overventilation typical for leaky construction. The Passivhaus concept is an exemplar of how far this model can lead in reducing heating demand. Whole house balanced supply and extract mechanical ventilation with heat recovery (MVHR) is an inherent part of such a strategy. In the UK MVHR is expected to be needed for highly energy efficient dwellings built to higher regulatory standards [1,2,3]. Continuous mechanical extract ventilation (MEV) is applied to air tight buildings that satisfy current energy efficiency requirements. Research has shown that in a maritime climate like UK, if embodied energy is included in calculation, the saving achieved with MVHR becomes marginal [4]. Climate change with expected milder winters is likely to further lower benefits of heat recovery [5,6]. In addition to this, there is an increasing need for heating and ventilation systems to exhibit greater resilience in the face of uncertainties relating to power supplies due to storms, flooding and resource depletion. Nevertheless as increased air tightness of new built constructions becomes normalised, discussion has moved towards health benefits for inhabitants mainly in relation to securing mechanical ventilation (MV) systems rather than any alternative [7,8].

Unfortunately, significant MV performance problems persist in relation to occupants. These include noise, poor access, lack of understanding of the system (e.g. thermal bypass switch, boost control) and lack of maintenance [11,12,13,14,15,16,17]. It has also been established that inhabitants tend to switch their MV off and end up with a reduced system performance. This scenario is regarded as major risk of under-ventilation in airtight dwellings. Therefore, gaining inhabitant's acceptance and intended interaction with MV becomes an important goal, particularly given UK 'zero-energy' goal for residential sector starting 2016. This paper proposes a framework linking factors that influence the emergence of the gap between design intentions and actual performance. The framework is based upon findings of previous studies as well as results of a one year-long in-depth Building Performance Evaluation of 40 households in two UK low carbon housing developments (Case A and Case B).

2. Emerging gap - process diagram

A theoretical process diagram is used here to identify the sequential stages in the housing lifecycle and linking factors that shape MV performance with the emergence of inhabitant practices related to ventilation (Fig.1). This lens is then used to examine multiple interdependencies with identified factors assigned to two distinct processes. One of the processes involves dwelling and the other one inhabitant. In principle the processes are distinctive - each happens within its own timeframe but some factors from the dwelling side strongly influence the flow of the process on the inhabitant side and vice versa. Such strong interdependencies are visually signaled in the diagram.

Each process is described as a sequence of stages (black circles) related to ventilation factors which either facilitate (progressive arrows) or hinder (returned arrows) the operation of MV as designed (shown by the central rectangle). It is the returned arrows which cause the emergence of the gap where subsequent factors are contingent upon precedent ones. The diagram highlights the problem of tackling only selected factors and of tackling them in the wrong order. Typically, user engagement in a well prepared learning process will not result in satisfaction with control over MV if a desired level of control is not possible due to design requirements.

2.1. Dwelling process

The initial dwelling design is shaped by the planning stages and context including client's expectations, setting goals, budget, designers' experiences etc. [18] These factors are not included here as the focus is on understanding how an already developed design intention (to secure good IAQ for a dwelling with continuous MV) influences the occupancy stage. Fabric design factors that have strong impact on ventilation design include air tightness, overheating risk and/or a dwelling's volume - these are typically based on assumptions about number of occupants and heating patterns [19]. Once continuous MV has been agreed, further decisions include specifying the system and its target performance (to achieve good air quality) linked with assumptions about costs and savings achieved in relation to ventilation. Future inhabitants need to know about these factors, if they're not involved in decision making process at design stage, in order to build up their expectation of the benefits related to using MV continuously [20]. The system specification also determines its vulnerability to faulty installation or commissioning

and availability of quality components on the market, affecting its resilience. A complex, emerging technology is more risky than a simple well established one [21]. A far reaching consequence of MV system specification is that it fixes the future capacity of inhabitant to interact with the system to adjust it to own needs. At one level the system capacity has to be fine-tuned for control and at another level there is the design intention related to the level of control available to the inhabitant. The fine tuning intention needs to be convincingly communicated to the inhabitant and reinforced through the design and location of controls. The feedback for inhabitants (its performance, energy use) is also decided at design stage. The outcome of design and procurement stages are 'as built' fabric and systems and their capacity to deliver as designed performance if used as intended. At occupancy stage the dwelling related barrier to as-design performance include events that were not covered by the design model, i.e. extreme weather events, power cuts, climate change, possibly also changes to site's direct context as well as product substitutions, re-positioning of construction elements or poor workmanship.

Fig. 1. Dwelling and user related factors hampering intended continuous MV operation in airtight dwelling.

2.2. Inhabitant process

The inhabitant process starts with any tacit knowledge about continuous MV for home ventilation. In this early transition period of building industry towards air tight housing in the UK, most inhabitants come from drafty homes

with no MV or intermittent one, usually linked to the light in a bathroom without a window. Here, the intuitive behaviour is to rely on natural ventilation, either passively based on air leaks or actively thorough windows opening. To change this intuitive approach to ventilation involves gaining awareness of having continuous MV installed and understanding the basic need for it. This is most likely to happen when the inhabitant is involved in shaping the design and makes a conscious decision about having the MV. Alternatively, when the inhabitant moves in the home handover process is used as the first step to raise this awareness and home user's guide to reinforce it. However, this opportunity for learning requires an initial intention to engage from the inhabitant. The Theory of planned behaviour suggests that environmental attitudes, self-efficacy or social pressure all play a role in triggering and enhancing learning intended to modify home use practices, including ventilation practices. A first a step here is for the inhabitant to consciously allow for continuous MV operation. Only then can experiences associated with this way of ventilating own dwelling start building up. These experiences may be both positive and negative. They may be built around actual observations (e.g. lack of condensation vs. noise) or expectations (e.g. indoor air quality or energy savings communicated by the designer vs. operational cost). Energy and cost related expectations may be difficult or even impossible for the inhabitant to prove. As long as they are turned into actual observation, assigning them to positive or negative category has to be based on inhabitant's trust and assumptions, which may be favourable or unfavourable towards MV, irrespective of reality. This perceived balance of positive and negative experiences is crucial for developing an inhabitant's judgment on the need for continuous MV as core ventilation strategy and is a critical step. Once the need for MV develops inhabitants can take substantial efforts to reduce the impact of negative issues experienced or even accept them. The effort may include learning to use the controls provided or go as far as checking the as-built performance and matching it with the design intention. On the other hand if the need for MV is not adopted, any issue experienced simply leads to inhabitants switching the system off and system failures are ignored. Even if daily operation of the MV system is fully automated the inhabitant still has to be convinced of the need for system's operation because inevitably maintenance is needed to sustain the desired performance and it is up to the inhabitant to make the effort to secure it.

2.3. Dwelling-inhabitant interaction.

Within the ventilation practice adopted by an inhabitant in a given home environment practices can be placed within three categories. Firstly, where the inhabitant relies on continuous MV and uses windows only as auxiliary ventilation, as designed for. Secondly, is a hybrid practice: switching between MV or NV on seasonal or diurnal basis, partly addressing design intentions. Thirdly MV can be ignored where the inhabitant relies on natural ventilation as in previous accommodation, ignoring design intentions altogether. The latter two options can still result in good indoor air quality when other non MV dependent factors are applied, i.e. openings design (cross ventilation, windows location and design) and window opening constraints. The constraints can be site specific (external noise level and air quality) or related to inhabitant (e.g. occupancy patterns, perceived safety, pets).

The lack of inhabitant's satisfaction with the achieved IAQ can be decisive in triggering a home use learning process and leads to switching between the three practices identified above.

If a link is identified between poor IAQ or high energy consumption and the MV operation is different than originally designed for, then the dwelling or inhabitant related factors leading to this gap need to be tackled. However, if despite the gap, the IAQ is good, energy consumption is within or below expectations and the user is satisfied, then it may be the design model or the assumptions behind it that need to be challenged.

3. Methodology

Comprehensive building performance evaluation methods used for this study (Table 1) identified ventilation practices of 105 households in the two case studies (Case A and Case B) in relation to design intentions, available means of control over ventilation, achieved satisfaction with control over ventilation and perception of internal environment across seasons. 40 of these households were covered in depth to examine the development stage of these practices, guided by theory of planned behaviour and practice theory. This included quantitative monitoring (24 July 2013 - 24 July 2014) to provide an objective physical performance baseline in relation to inhabitant's subjective responses (Table 1). Adaptive thermal comfort theory shaped project design and the alliesthesia concept

also helped with the analysis [19]. Action research included interim feedback (reports) and informing inhabitants about health risks observed with discussion through meetings of research findings. This helped identify evolving ventilation practices as a result of increasing inhabitants' understanding of MV. It also led to the recommissioning of the MVHR system in Case A in May 2014.

Table 1 Critical factors and methods used in evaluation strategy for the 40 dwellings. Ventilation related quantitative/qualitative factors Research Methods

Dwelling

Environmental design goals

Ventilation in design and procurement process

Supply chain/workmanship issues

Fabric and ventilation systems as designed/as built:

MV design, specification, installation and commissioning, air tightness, overheating risk

NV: opening's design - cross ventilation, site related window opening constraints

Scope of intended user control over MV, MV

Air flow: compliance with building regulations

inhabitant's complaints vs. issues identified

Noise: MVHR operation against background noise

IAQ average + issues: overheating, increased RH levels, CO2 above 1000ppm

Robust link of energy consumption and ventilation practices adopted

Inhabitant

Previous accommodation (experience with air tight homes & continuous MV)

Initial awareness of ventilation system installed Engagement in design/ procurement

Accuracy and coherence of information given

Perceived usefulness of this stage

Engagement in ventilation related learning

Perception of control over MV and individual comfort range (satisfaction against temp. monitoring)

Understanding & skills to interact with MV controls

Prevailing occupancy patterns (windows opening)

Ventilation practices: continuous MV with axuxilary NV, hybrid or only MV, behavioural change observed

A case study approach is used to analyse ventilation practices of households. The details of the two case studies are described inTable 2.

Table 2 Case study characteristics.

Case study Case study A (20 participants - 100%) Case study B (18-20 participants - ca. 10% sample of

occupied units)

Completion 2013 2011

Size + units Mutually owned 20 units: 8 houses (3&4 bed), 234 units: 1&2 bedroom

12 flats (1&2 bed) Owned/shared ownership/rented

Interview with design team Environmental ratings achieved

Construction audit - on site against design documents

Commissioning check, SAP check, airtightness certificates, walk through + photographic survey

Usability survey

MV air flow check (5 dwellings) + shadowing MVHR recommissioning MVHR acoustic check (3 dwellings)

Temp., RH monitoring and CO2 (4 dwellings for a year + 5 dwellings for 4 moths) monitoring

Gas & electricity meter readings

Extended BUS survey (n=105), interview Interview with residents, design team & client

Shadowing of the introduction of occupants to their home (Case study A only) + evaluation of home user guide (HUG) & manuals, Interview

Extended BUS survey (n=105), Interview Usability survey

Interview & repeated home visits every 7-8 weeks Walk through, home visits every 8-9 weeks

Extended BUS survey (n=105), Interview, Temp, RH and CO2 monitoring

Dwelling types

New build terrace, semi-detached houses, apartments - cross-ventilation

Houses:2; Apartment block:3

Designed: q50=4-5 m3/hr.m2

As-built: q50=1.42-4.3 m3/hr.m2

gas and electricity + renewables on site

MVHR: unit Vent Axia Sentinel Kinetic

Refurbishment 1950's apartment block: single aspect (east or west facing)

Designed: q50=7 m3/hr.m2

As-built: q50=4.29-5.33m3/hr.m2 (4 cert.)

electricity

MEV: fans - Greenwood Unity CV100

2006 UK Bld. Reg. (retrofit) + Eco Homes Very Good

No. of floors

Air permeability

Energy Ventilation

Energy standards Code for Sustainable Homes Level 4

4. Analysis

4.1. Dwelling: design, procurement, occupancy

Mechanical ventilation design for both developments is based on continuous fans operation and includes humidistats to trigger automated boost in case of increased RH. Manufacturers of both types of fan units state that their products to be 'ultra-quiet'. These claims are based on standarised laboratory tests which do not account for the noise component of air flow through ducting. Noise levels from 'as built' systems vary between dwellings (sound level measurement in 4 dwellings) and in some it was perceived as a significant nuisance by the inhabitants. In both cases the intended user control of MV is restricted mainly to a manual boost switch. In Case Study A, in order to allow safe access to MVHR filters cleaning the mains switch is in an exposed location. The MVHR unit control panel, intended mainly for commissioning and servicing purposes, is not readily available to the user. All inhabitants are aware of this panel but few attempted to use it. Equally, in Case Study B the mains switches for fans were not intended to be used and are 'hidden' in an inaccessible location high in the utility cupboard. Ironically, residents who discovered their function actually found them useful for control purposes and complained about poor access.

A major difference between the two Case Studies is the designed natural ventilation component: efficient cross-ventilation in Case A and single aspect with no cross ventilation in Case B. Interestingly MVHR was introduced into the design of Case A after the architectural design was finished - active ventilation through windows opening was the initial ventilation strategy. In Case B the lack of cross ventilation and continuous extract ventilation were both planned from the start. In both Cases installation and commissioning issues were identified linked with in use issues such as noise or under-ventilation. How well the inhabitant ventilation practices match the above design strategies is described in the next section.

Procurement and commissioning issues related to MV were identified in the course of research for both case studies. Design and commissioning documents audit, walk through, feedback from inhabitants, air flow rate check (performed in 4 Case A dwellings and 1 Case B dwelling) and noise level check (3 dwellings in Case A) resulted in understanding the gap between designed and as built performance. Design issues in Case A related to diffusers disposition were picked up and the MVHR system was recommissioned in all dwellings with ceilings taken down in kitchen areas in order to insulate external air supply ducts. The works caused major disruption for the occupants but all agreed to go through with the process in order to improve the MVHR performance.

4.2. Inhabitant: tacit knowledge, awareness raising and adoption of new practices

None of the participants in either Case Study had prior experience with energy efficient air-tight dwellings. All came from traditional houses or flats and all but one without continuous MV. One inhabitant in Case A had MVHR installed in his previous house for environmental reasons. In the interview he perceived this prior experience as an advantage: 'A lot of people think 'Oh, you've got to open up the windows'. They can't think of that [MVHR] as a source of fresh air. Whereas to me I just take it for granted.'

All inhabitants in Case A were involved in design or procurement stage, so everyone was aware of having MVHR installed as a part of mutually accepted energy efficiency strategy. Concise written advice from contractor was circulated instructing inhabitants to keep the system permanently on in order to save energy by recovering heat.

Advice was also given to keep the windows closed to increase system's efficiency. Moving-in coincided with a cold spell - everyone started using heating and had the MVHR on as advised. No one therefore questioned the need for having the system on when heating the house. However, inhabitants who were comfortable with lower temperatures (annual mean temp. in the coldest bathroom: 16.6oC) and didn't use the heating but suffered from some performance issues started temporarily switching the system off. Through visits and interviews it was established that noise affected 8 households, draft - 1 household, condensation leaking from the unit - 3 households. In 6 households, noise was regarded as most annoying in the evening when going to sleep. Technically skilled and inquisitive inhabitants discovered through the MVHR installation manuals how to program low settings for selected periods of the day. This functionality was not intended for inhabitants, however; in order to access it the MVHR unit needs to be restarted and accessed in service mode. However in 4 dwellings the low setting was intrepidly programmed for two night hours which solved the noise problem when trying to go to sleep. In one household the inhabitants remained unaware of low settings option and simply switched the system off altogether. 3 households accepted the noise. The availability of cross-ventilation combined with environmentally driven urges to switch off all energy consuming appliances prompted the majority of inhabitants to develop hybrid ventilation practices. The anxiety about MVHR related energy consumption was deepened by the inability to check it in Case A - manuals did not prove to be helpful even for technically advanced inhabitants. All but 2 households opened their windows for ventilation as a result. Preference for keeping the windows open (often linked with switching MVHR off) in favourable weather when at home was expressed in 90% of the interviews. This ties in with findings from previous research [22,23,24,25]. Monitoring conducted did not indicate worse IAQ in the identified heatwave when the MVHR systems were most likely to be off (RH + CO2 in 3 dwellings). Similar findings were established by Brown et al [14]. RH exceeding regulatory thresholds was observed in households that relied solely on window opening in cold and wet months [26].

In Case Study B inhabitants moved into finished dwellings going through structured home handover procedure and were presented with a home user's guide covering all aspects of development, including ventilation. In Building Use Studies (BUS) survey, out of 95 questionnaires returned, 19 inhabitants stated they did not have MV installed or did not know what it was. A further 8 respondents left all ventilation related questions blank which may also suggest lack of awareness of the subject. Up to 30% of inhabitants lacked basic awareness of having the MV system. This could be interpreted as result of 'ultra-quiet' fans working unnoticed in the background. This would be acceptable only if maintenance or repairs managed by inhabitant weren't a necessity to avoid drastic drop in the fans' performance [...]. The high number of inhabitants unaware of having exhaust fans may partially be an unintended result of the fan being disguised with a firmly screwed down aesthetic cover. Difficult access and lack of visibility of the grill may also explain why none of participants were aware of the need to clean it. 4 BUS respondents admitted their fans were malfunctioning from the start - they never solved the issue but at least they knew they had the fans. A further 8 inhabitants among those who are aware of having the MV claimed they had never felt the need to use MV. In Case Study B the ventilation related learning process in over 40% of households never got to the stage of inhabitants trying to use MV as intended in order to test it. The following participant's quote illustrates the impact of previous experience with automatically varied MV, when the learning process in the new home does not go beyond awareness of having the fans installed: 'I find it [MEV] quite ineffective and it is not automated so I have to keep reaching for the mains switch.' 17 inhabitants admit intermittent use, mostly when showering. Among many reasons for switching the fans off, energy saving is primary (38 responses), noise is second (29) followed by heat loss (14). Anxiety about fans energy consumption expressed by majority of inhabitants is much less justified in summer conditions in case of local MV fans than in the case of MVHR system [9]. Fans specified in Case study B only consume ca. 2W whereas the MVHR operation consumes 20-80W. Informing 4 Case B participants of the in depth study, who indicated running cost as the main reason for keeping the fans off, about actual energy consumption by fans resulted in changing ventilation practices by 3 of them. In terms of noise there was no way for Case Study B inhabitants to reduce it, unlike in Case Study A. During home visits and in interviews, noise from fans was described as most disturbing when going to sleep, relaxing or having a bath. No one linked it with taking a shower. This association with specific activities requiring silence is similar in both Case Studies and is highly relevant for our conclusions.

4.3. Inhabitant: ventilation practices.

Case study A n=20

When do you have mechanical

ventilation system switched on?

When do you have your windows open?

Case study B n=95

When do you have mechanical

ventilation system switched on?

When do you have your windows open?

Winter Spring Summer Autumn Design intention

Winter Spring Summer Autumn

Winter Spring Summer Autumn Design intention

Winter Spring Summer Autumn

I Never Orly when at home ■ Other ■ All the time

I Never Only when at home Other All the time

Fig.2 Seasonal variation in declared operation of MV and window opening (BUS survey).

A BUS survey carried out in Feb 2014 was extended by authors with questions focused on MV operation and windows opening practices. All households in Case Study A claimed to use MVHR, though only 25% used it continuously, as intended by designers. Significant variation between seasons is due to over half of households switching MVHR off correlated with the opening of windows in warmer seasons (Fig.2). In winter all but one household have the MVHR continuously on and 70% never open the windows. This indicates that 75% of Case A households developed hybrid ventilation practices relying on MV or NV depending on the weather or time of day. This finding was confirmed with home visits and interviews. Diurnal variation was motivated by noise and windows opening that for many meant no need for MV operation. Outside noise prompted closing the windows for the night and switching the MVHR on even during the heatwave (1 household). Noise caused by MVHR operation caused switching the MVHR off regardless of the season. Windows opening was explained by the need to provide 'fresh air', audible connection with the outside (birds, leaves, social life), coping with excessive heat and habit.

In Case Study B only 8 households out of 95 (less than 10%) claim to have all extract fans (kitchen, bathroom) operating continuously throughout the year, even though this is designed as essential. The worrying finding is that there is no direct correlation between MV operation (similar across the seasons) and windows opening (more windows open in summer). Winter windows' opening is similar in both developments however almost half of inhabitants in Case B never turn their MV on, including winter (Fig 2). This indicates that in Case B almost half of household ignore MV and rely on natural ventilation switching between active (windows opening) or passive (air leaks). Only 4 airtightness test certificates were available for the study, including 1 for the participating apartments. These show permeability within the range 4.29-5.33m3/hr .m2. Home visits and thermal imaging allowed identifying significant air leaks around windows (2 dwellings) or through gap under the front door (6 dwellings) that allows some continuous air change - which is good given that the fans are off and windows closed. However in 3 out of 20 dwellings there were no complaints about drafts from the windows, no trickle vents were installed, draft excluder in front door is dropping, fans are off and windows closed in cold weather. Monitoring indoor air quality was limited to dry bulb temp. and RH measurements taken every half an hour in three locations in each dwelling for a year. CO2 monitoring was performed in one living room area (equipped with trickle vents) for a year and in 5 bedrooms for 4 months. Results are presented in another forthcoming paper. Generally they indicate that there is no clear cut difference between those few who use MV as designed i.e. continuously and others. No one relies on MV only. All households open windows in the summer due to severe overheating problem in most dwellings. In terms of exceeding RH maximum thresholds given in building regulation's, it has been identified only when a few factors coincide: significant amount of moisture is released (through showers, drying clothes) all air supply points are sealed, fans are off and use of heating is very limited or none.

5. Discussion

The broad scope of research allowed capturing the sequence of factors that have led to the identified gap between design and in use performance and operation of continuous MV identified in two developments. The two case studies represent different typologies (community led small vs. developer led large), community types (intentional community vs. mostly anonymous) and demographics (all ages and family situations vs. mainly young working singles or couples). Only Case A represents participatory design that covered conscious decision on having MVHR installed. MV systems in the two Cases represent different complexity level: whole house balanced with heat recovery vs. local extract fans. Also handover process varied between the two: all households in Case B vs. some in Case A. Maintenance is managed by the community in Case A and by large residential managing company in Case B. There are however also similarities among the 'gap factors' between the two case studies:

• intended scope of daily user control over MV is similar: limited to manual boost button.

• automated control linked to increased RH level is included

• procurement and commissioning issues identified in both developments as a result of BPE

• lack of previous experience with air tight dwellings and continuous MV is prevailing

• noise issues are experienced in some dwellings, disturbing in particular when going to sleep

• inhabitants are concerned about energy use (mainly environmental reasons in Case A, bills in Case B)

• energy use/operation cost of MV is not clear to all but one inhabitant across the two Cases

These factors have led on average to significantly different results in terms of including MV into households' ventilation strategy. In Case A everyone knows they have MV and use it at some point: permanently in winter. Performance issues became apparent in some households and they have been gradually tackled as far as it was possible. Improving and adjusting performance to inhabitants needs was done despite serious disturbance it caused and included exercising control over the system intended by the designer for service reasons only. Few rely on MV only and hybrid ventilation practice is prevailing. In Case B however MV has never been tested by many inhabitants unaware of having it or not feeling the need for it. Initial home use learning proves to be inefficient. As it has been always off performance issues were not experienced. Among those who did try to use it and experienced for example noise, the system is permanently switched off or used intermittently; only when an inhabitant sees the purpose of it. There is no way to control the system other than to switch it off.

Interestingly, the monitored IAQ doesn't show a clear-cut difference between those who ventilate their homes as designed and others - some of those who have adopted other ventilation strategies still achieve good IAQ and high satisfaction level. This suggests that design assumptions in relation to continuous MV in housing can be challenged.

6. Conclusions

Tackling the gap between design intention and actual performance and operation of continuous MV in housing context is a 'wicked' problem [27]. Understanding the need to approach it in a certain order is vital to achieve real impact. Hierarchical dependencies related to dwelling and inhabitant shown here indicate that practices are based on experiences from previous accommodation that can be modified through learning and building understanding of the need to change one's own ventilation strategy. Without understanding and accepting the need for change, old ventilation practices stay even if mould appears. Equally, securing a steady minimum air flow does not account for inhabitants' strong need for variation in air flow: 'fresh air' associated with window opening. Similarly, raising inhabitant's environmental awareness brings a need for more transparent and clearly justified energy consumption resulting from MV operation. Existing justifications for MV systems [28] must be questioned when wider then assumed temperature comfort ranges are accepted by the inhabitants. Two categories of factors hampering occupant use of MV as designed emerged from the study: those specifically related to the industry and user transition period towards low energy buildings and permanent adjustments that are required to the design of the system.

The rate of successful interaction with mechanical ventilation in dwellings can be significantly increased if the learning process is better supported and user's varied expectations are met in terms of control over the system. One important area identified for improvement from this study relates to the continuous MV model supported by current regulations. Modifying this model by making allowances for interrupted ventilation strategies which nevertheless maintain IAQ would allow for a diurnal quiet period to aid sleeping and avoid noise 'nuisance' at this time.

Additionally, a hybrid ventilation model [29,30] that allows for seasonal modifications and MV 'sleep mode' in case of CO2 levels below a certain threshold (eg. 500ppm) would allow an effective natural ventilation contribution. These findings significantly challenge existing MV design assumptions.

Acknowledgements

This work was supported by EU funding through Marie Curie IEF (BuPESA No. PIEF-GA-2012-329258). References

[1] DCLG. Government response to the Environmental Audit Committee Report: Code for Sustainable Homes and the Housing Standards Review; 2014. Available at: www.gov.uk/government/publications

[2] DCLG. Code for Sustainable Homes Technical Guide, Code Addendum; 2014. Available at: www.gov.uk/government/publications

[3] NHBC. Indoor air quality in highly energy efficient homes - a review. NHBC Foundation; 2012.

[4] Hernandez P, Kenny P. Integrating occupant preference and life cycle energy evaluation: a simplified method. Buildings Research & Information 2010;38(6):625-637.

[5] Frank T. Climate change impact on building heating and cooling energy demand in Switzerland. Energy & Buildings 2005;37:1175-1185.

[6] Sassi P. A natural ventilation alternative to the Passivhaus standard for a mild maritime climate. Buildings 2013;3:61-78.

[7] Howieson SG, Lawson A. et al. Domestic ventilation rates, indoor humidity and dust mite allergens: are our homes causing the asthma pandemic? Building Services Engineering Research and Technology 2003;24(3):137-147.

[8] Hasselaar E. Health risk associated with passive houses: An exploration. In: Proceedings of the 11th International Conference on indoor air quality and climate, Copenhagen, Denmark 2008, http://repository.tudelft.nlView/ir/uuid:88fd72b2-f7ab-45ea-a403-ce367801cf3f/

[9] NHBC. Assessment of MVHR systems and air quality in zero carbon homes. NHBC Foundation; 2013.

[10] Yoshino H, Murakami S. et al. Survey on minimum ventilation rate of residential buildings in fifteen countries. Proceedings of the 25th AIVC Conference-Ventilation and retrofitting, Prague; 2004:227-238.

[11] Balvers J, Bogers R. et al. Mechanical ventilation in recently built Dutch homes: technical shortcomings, possibilities for improvement, perceived indoor environment and health effects, Architectural Science Review 2012;55(1):4-14.

[12] Stevenson F, Fewson K. et al. Lancaster Co-housing Project Part 8: Post-Occupation Building Performance Evaluation, Green Building 2013; 23(1):24-35.

[13] Macintosh A, Steemers K. Ventilation strategies for urban housing: lessons from a PoE case study, Building Res. & Inf. 2005;33(1):17-31.

[14] Brown C, Gorgolewski M. Understanding the role of inhabitants in innovative mechanical ventilation strategies, Building Research & Information 2015;43(2):210-221.

[15] Siddall M. Problems in residential design for ventilation and noise, Green Building 2014;3:52-56.

[16] Derbez M, Berthineau B. et al. Indoor air quality and comfort in seven newly built, energy-efficient houses in France, Building and Environment 2014;72:173-187.

[17] Larsen TS, Jensen RL, Daniels O. The Comfort Houses: Measurements And Analysis Of The Indoor Environment And Energy Consumption In 8 Passive Houses 2008-2011, Aalborg: Department of Civil Engineering, Aalborg University; 2012, (DCE Technical Reports; No.145). Retrieved from: http://forskningsbasen.deff.dk/Share.external?sp=S6316263b-ab10-4c58-a899-f97cdcbc179f&sp=Saau

[18] Sinclair, D. (ed.) Plan of Work 2013. Overview. 2013, RIBA, London. Available at: www.ribaplanofwork.com

[19] CIBSE KS17 Indoor Air Quality and Ventilation, 2011, CIBSE. http://www.cibse.org

[20] Brown, Z. & Cole, R.J. Influence of occupants' knowledge on comfort expectations and behaviour. Build. Res. & Inf. 2009;37(3):227-245.

[21] Gorak, S Robust Technologies. Batiment International, Building Research and Practice 1990;18(3):162-168.

[22] Parkinson T, de Dear R. Thermal pleasure in built environments: physiology of alliesthesia, Build. Res. & Inf. 2015;43(3):288-301.

[23] Healy S. Air conditioning and the 'homogenization' of people and the built environment, Build. Res. & Inf. 2008;36(4):312-322.

[24] Yun Y, Steemers K, Baker N. Natural ventilation in practice: linking facade design, thermal performance, occupant perception and control, Building Research & Information 2008;36(6):608-624.

[25] Frontczak M, Andersen RV, Wargocki P. Questionnaire survey on factors influencing comfort with indoor environmental quality in Danish housing, Building and Environment 2012;50:56-64.

[26] Baborska-Narozny M, Stevenson F, Chatterton P. A Social Learning Tool - barriers and opportunities for collective occupant learning in low carbon housing, Energy Procedia 2014;62:492-501.

[27] Rittel HWJ, Webber MM. Dilemmas in a general theory of planning, Policy Sciences 1973;4:155-169.

[28] Lowe RJ. Ventilation strategy, energy use and CO2 emissions in dwellings - a theoretical approach, Building Services, Engineering, Research & Technology 2000;21(3):179-186.

[29] William J.N. Turner, Iain S. Walker, Using a ventilation controller to optimize residential passive ventilation for energy and indoor air quality, Building and Environment 2013;70:20-30.

[30] Sherman MH, Walker I. Meeting residential ventilation standards through dynamic control of ventilation systems, LBNL; 2011.