See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305619149

Unintended consequences of sustainable architecture: Evaluating overheating

risks in new dwellings

Conference Paper · July 2016

CITATIONS READS

15 246

3 authors, including:

Linda Toledo Andrew John Wright

Eurac Research De Montfort University
11 PUBLICATIONS   87 CITATIONS    65 PUBLICATIONS   1,419 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

THERM - THrough life Energy and Resource Modelling View project

co-design for sustainable housing in Peru View project

All content following this page was uploaded by Linda Toledo on 25 July 2016.

The user has requested enhancement of the downloaded file.

 PLEA 2016 Los Angeles - 32th International Conference on Passive and Low Energy Architecture.
Cities, Buildings, People: Towards Regenerative Environments

Unintended consequences of sustainable architecture:

Evaluating overheating risks in new dwellings

LINDA TOLEDO1, PAUL C CROPPER1 AND ANDREW J WRIGHT1

1
Institute of Energy and Sustainable Development, De Montfort University, Leicester, UK

ABSTRACT: Governmental strategies to reduce heating demand from dwellings have led to a range of problems
relating to ventilation and occupant comfort. In fact, growing evidence of uncomfortably warm homes has been
appearing in UK consistently in the few last years. This paper discusses the overheating risk in four highly insulated
homes in the UK where a mixed methods approach has been deployed to characterise areas of overheating risk, which
have been found to occur with different degree of severity and different sources of risk, all related to design and
occupant behaviour.

Keywords: thermal comfort, overheating, residential mixed-mode, field studies, Passivhaus, low carbon design

INTRODUCTION 2012; Morgan, Foster, Sharpe, & Poston, 2015;

Today’s concern about climate change and its Tabatabaei Sameni, Gaterell, Montazami, & Ahmed,
consequential humanitarian impact has led the UK 2015). This study is a contribution to that tradition, as it
government to develop strategies to reduce greenhouse also investigates the design choices and consequences
gases emissions (Crown, 2008; HM Government, 2011). that govern the thermal strategy of highly insulated
In an attempt to reduce energy consumption and homes in the UK, by focusing, more specifically on
associated carbon emissions from buildings, substantial current building regulation standards and Passivhaus-
changes have recently been made to UK building like buildings.
regulations, resulting in homes with significantly The study of energy efficiency and thermal
improved standards of thermal insulation and much comfort in highly insulated dwellings by its own nature
higher levels of airtightness (DCLG, 2013). stands between the spheres of building physics and
However, growing evidence of uncomfortable social science. Accordingly, this paper presents initial
(overheated) new energy efficient dwellings in the UK results of a mixed methods research strategy that takes
has appeared in the literature (DCLG, 2012; Dengel & into account the transdisciplinary nature of the discipline
Swainson, 2012; Taylor, 2014). This appears to be an of architecture, by integrating real world quantitative
unintended consequence of the UK CO2 mitigation and qualitative data collected from building performance
agenda (Davies & Oreszczyn, 2012) characterised by the evaluation and interviews.
dichotomy between highly efficient dwellings and
summer thermal comfort. METHODS
Thermal modelling offers a powerful tool to The context of this paper is a larger study aimed
predicting the possibility or probability of overheating at determining the likelihood of overheating in highly
and can be used to test the consequences of changes in insulated dwellings in the UK, the sources of
specific parameters, such as orientation, house types, overheating risk, and the relations between overheating
house layout, climate change, etc. under well-defined risk, on the one hand; and building design and occupant
conditions. However, it has become clear that thermal behaviour, on the other hand.
modelling studies are not able reliably to model human In order to achieve this objective, an in-depth
behaviour and their thermal interaction with their study has been performed on four highly insulated
environment (Beizaee, Lomas, & Firth, 2013) as these British homes, where data has been collected by (a)
introduce unknown variants to parameters such as conducting observational surveys on site, (b) recording
ventilation rate. Therefore, the design of energy efficient environmental parameters and (c) submitting a number
dwellings needs to be informed by knowledge gained of questionnaires to the occupants. These are intended to
from newly built highly insulated dwellings, leading to a evaluate the physical environmental measurements as
built environment sustainable for people’s needs and well as the occupants’ and design’s role in the thermal
resilient to the changing climate. performance of these homes.
Monitoring studies of dwellings with perceived The homes are of different types, which are
overheating have often collected information about both presented in Table 1 and Table 2. None of the houses
the use and the construction of the building (DCLG, made use of any cooling devices such as fans or air
 PLEA 2016 Los Angeles - 32th International Conference on Passive and Low Energy Architecture.
Cities, Buildings, People: Towards Regenerative Environments

conditioning units. It is worth noting that UK51 was the tenant relates to the environmental measurements in
only refurbished (19th century) house, to a very high their home. The physical environmental monitoring was
near-Passivhaus standard thermally. performed both by means of high-resolution intervals
measurement and through spot measurements. The
Table 1: Overview of case studies homes with main intervals measurement of air temperature (˚C) and
construction characteristics. Relative Humidity were recorded every 10 minutes from
30 June 2015 till 13 August 2015. During this period a
House House type U-value Thermal Orien- short heatwave occurred in England (from the 30 June
code n. bedrooms ext. walls mass tation 2015 till 2 July 2015), with temperatures exceeding
& location (W/m2.K) 30˚C on 1st July 2015.
UK51 Refurbished 0.12 NO E-W Overall, the occupier questionnaires were
terrace administered in order to collect information about the
2 bedrooms
Leicester (UK)
house, its occupancy and the interactions between
UK52 New 0.09 NO N-S occupiers and the house: collection of background
detached information about the house (microclimate, physical
bungalow dimensions, occupants’ background and household
2 bedrooms composition) was integrated with information about
Sandiacre (UK) what the occupants think of their thermal environment
UK54 New 0.19 YES N-S and how the occupants adapt/interact with their thermal
terrace environment. The questionnaire included both ranking
3 bedrooms questions and qualitative open questions in order to
York (UK)
determine if there were any problems with the design
UK55 New 0.19 YES E-W
that is not accounted for by the ‘forgiveness’ factor (by
detached
4 bedrooms forgiveness factor, it is meant the fact that “occupants
York (UK) tolerate less than perfect conditions because they like
the overall feel and design quality of a building”
(Nicholls, 2008, p.282).
Table 2: Overview of ventilation and solar control availability
in the studied homes. RESULTS
Temperatures plots
House Ventilation Cross Solar gain control Plotting the internal temperatures against the
code type ventilation former CIBSE overheating criteria (2006) where a
UK51 MVHR YES Internal blinds threshold of 26˚C and 28˚C is defined for bedrooms and
(partially) living rooms respectively, made it evident that most of
UK52 MVHR YES NO the high temperatures were located in the bedrooms on
the upper floors. The living rooms performed better in
UK54 MV YES external overhangs
(partially) terms of summer comfort. In fact, living rooms in
UK55 MVHR YES external overhangs houses UK51 and UK52 exceeded 28˚C threshold only
(partially) during the short heatwave experienced in the UK that
summer.
More specifically, the temperature-related
Longitudinal data were collected through HOBO conditions of bedrooms and living rooms in the houses
loggers, which recorded environmental parameters, such under review can be described in terms of (a) mean
as air temperature and Relative Humidity. Loggers were temperatures, (b) minimum temperatures and (c)
placed in every habitable room and the results have been maximum temperatures and temperature variation, as
analysed to map the most problematic areas or rooms in follows and as shown in figures 1, 2 and table 3:
the house. In addition, the main bedroom temperatures (a) Mean temperatures: mean temperatures in the
were evaluated according to the CIBSE overheating living rooms were lower than those of the bedrooms in
threshold approach (CIBSE, 2013). all houses, mostly under 23°C. In the bedrooms, mean
These longitudinal measurements were temperatures were up to 2°C higher. In house UK51, the
complemented by a number of occupant questionnaires mean temperatures recorded in the bedrooms were
for post-occupancy evaluation. Questionnaires were between 24°C and 25°C. In house UK52, the mean
aimed at collecting a feedback on the effectiveness of temperature recorded in the bedroom was just below
new highly efficient designs, as well as collecting data 24°C. In house UK54, the mean temperature recorded in
regarding occupants’ behaviour, occupants’ control, and the bedroom was below 23°C. In house UK55, the mean
occupants’ thermal comfort sensation in order to capture temperature recorded in the bedroom was just below
how perceived thermal comfort and behaviour of the 25°C.
 PLEA 2016 Los Angeles - 32th International Conference on Passive and Low Energy Architecture.
Cities, Buildings, People: Towards Regenerative Environments

(b) Minimum temperatures: in house UK51 10/13°C. In house UK54, the mean temperature
minimum temperatures in all the rooms considered were recorded was 22.5°C and 23°C in the living room and
above 21°C at all times. In house UK52 the minimum bedroom respectively. Noticeably, this house (UK54)
temperatures recorded were lower (16-19°C). Minimum presented the lowest maximum temperatures among all
temperatures in house UK54 were just below 21°C, but, the case studies. Also, the rooms presented the smallest
in sharp contrast with the other houses, temperatures variation in temperatures of 6 K. In house UK55, the
have been maintained with no high peaks. Instead, in first floor bedroom recorded a maximum 31°C whereas
house UK55, the minimum temperatures of dining room the ground floor dining room recorded 27.3°C. In
and bedroom were below 17°C and above 21°C addition, it was noted that the temperatures recorded in
respectively. This reveals a different management of the the sunspace (winter garden) were extremely high,
temperatures within the house, since the dining room is swinging from a min. 20°C to a max, 42°C. This could
located next to the kitchen and some extra heat gain be attributed to the lack of both ventilation and solar
could be expected to contribute to the temperatures. control; and this can be confirmed by this sunspace’s
orientation (East), the lack of solar shading and by the
34
occupant not opening the windows, resulting in heat
32 gains then released in the adjacent rooms (see fig. 3).
30
28 43
wintergarden
26 38
24 adjacent living room
22 33
20
28
18
16 23
18

Figure 1: Max, min. and mean temperatures (°C) for living

areas.

34
Figure 3: Recorded temperatures (°C) in the sunspace and
32
adjacent living room from 30 June 2015 till 7 July 2015.
30
28
26 It is noticeable the fact that whilst UK52 and
24 UK54 have almost similar average temperatures, they
22 have at the same time a remarkable difference in
20 temperature variation and maximum temperatures.
18 These two houses were designed to optimize the use of
16 natural ventilation through the windows; however house
UK52 has a much higher level of insulation and it has
no thermal mass exposed.

Table 3: Max, min., average and range temperatures

Figure.2: Max, min. and mean temperatures (°C) for Temp °C Temp °C Temp °C Temp K
bedrooms. min max mean range
UK51-bed 1 21.5 29.3 24.3 7.8
(c) Maximum temperatures and temperature UK51-bed 2 21.2 33.7 24.8 12.6
variation: house UK51 recorded a max. of 29°C in the UK51-living 21.5 28.5 23.7 7.0
first bedroom and 33.7°C in the second bedroom . In UK52-bed 1 19.2 30.1 23.6 10.9
the second bedroom the temperatures difference is 12.6 UK52-living 16.7 30.3 22.3 13.5
K, presumably due to natural ventilation. This was later UK54-bed 1 20.7 27.0 23.0 6.2
confirmed by the occupants, who also claimed that they UK54-living 20.6 26.7 22.5 6.1
found this room uninhabitable during the heatwave. In
UK55-bed 1 21.6 31.2 24.9 9.6
house UK52, all occupied rooms showed a maximum of
UK55-dining 16.4 27.3 23.0 10.8
30°C with temperatures showing a higher variation of
 PLEA 2016 Los Angeles - 32th International Conference on Passive and Low Energy Architecture.
Cities, Buildings, People: Towards Regenerative Environments

Frequency distribution and histograms in late spring/early summer 2015 and the second in mid-
The former CIBSE Guide A (CIBSE, 2006) summer 2015.
provides guidance for indoor temperatures for buildings When asked how often the windows were kept
including summer indoor comfort temperatures. The risk open in order to cool a room, occupants from houses
of overheating is calculated by thermal performance of UK51, UK54 said that they left the windows open to
buildings, measured against both a benchmark cool the house day and night. By contrast, the occupant
temperature and an overheating criterion. The general of house UK52 left the windows open during daytime
indoor comfortable operative temperature for bedrooms only, due to security concerns. Lastly, no cooling
as listed in CIBSE Guide A is 23°C and the threshold through windows opening was used in house UK55.
operative temperature is 26°C. This overheating criteria These behaviours seemed confirmed by the temperature
establishes that if the mentioned threshold temperature readings of most houses; the only exception is house
(26°C) exceeds 1% of the occupied hours, the building UK51, where opening the windows seems to be
suffers from overheating (CIBSE, 2006). It is worth insufficient to properly cool the house. These opinions
mentioning that this criterion is been revised to embed where mostly maintained in both the first and second
the adaptive approach (CIBSE, 2015). questionnaires.
Looking at the frequency distributions (figure 4), When asked how difficult is it to keep
and although in different intensity, all bedrooms comfortably cool, the responses from early till late
exceeded the 26°C threshold. House UK51 and UK55 summer (i.e. after the heatwave) showed that the
showed the most severe cases of high internal occupants of house UK51 have difficulties in keeping
temperatures. In fact, when comparing the internal cool the second bedroom (under the roof). For this
temperatures amongst a threshold of 26°C, it was found reason, the occupants of this house slept in the cooler
that house UK55 presented the worst performance with living room during the heatwave. The occupant of house
a 23% of monitored hours were above the 26°C UK52 claimed that she did not have a similar difficulty,
threshold. This was followed by house UK51, where in early summer. However by mid-summer the occupant
bedrooms 1 and 2 showed respectively an 8% and 15% did find it difficult to sleep due to excess heat and opted
percent of monitored hour above threshold. House to go to the living room to open a window and - concern
UK52, showed a 10% over threshold hours. Finally the of burglary- keep herself awake by reading a book. The
best performing house in terms of overheating was occupants of house UK54 said that they had difficulty to
house UK54, where the threshold was exceeded for only keep cool only in the small office, where windows were
the 2% of monitored hours. not opened due to building works nearby. Finally, the
occupants of house UK55 stated that they experienced
no difficulty in maintaining room temperatures
comfortably cool in the first questionnaire; however in
the second questionnaire the same occupants claimed
that they had difficulties on keeping comfortably cool
temperatures throughout the whole house and during all
day, to the point that they felt the necessity to go outside
in order to feel thermal relief.

DISCUSSION
Looking at the bar charts and the hours “above
26°C” (CIBSE overheating threshold), one can notice
that the worst performing bedroom (in UK55) and the
best performing bedroom (in UK54) are located in the
same development and have the same materials and
building specifications. The difference in those cases
Figure 4: Stacked bar charts of temperature ranges of may partly be explained by the different designs and
bedrooms orientation, different ventilation system and ventilation
management. In fact, whereas house UK55 delegates the
Questionnaires provision of thermal comfort to the MVHR system, the
Occupants responded to two identical occupiers of house UK54 managed ventilation manually,
questionnaires, one submitted at the beginning of the thus allowing to keep the heat out during the day and to
study and one at the end of it. The goal of this strategy ventilate during the night. In a similar vein the two
(double questionnaire) was that of seeking confirmation houses that rely most in natural ventilation had a
of the occupants’ opinions and behaviours throughout reduced number of hours above the 26 °C threshold in
different seasons. The first questionnaire was submitted this group. By contrast, the houses that manage
ventilation through MVHR presented the highest
 PLEA 2016 Los Angeles - 32th International Conference on Passive and Low Energy Architecture.
Cities, Buildings, People: Towards Regenerative Environments

number of hours above the 26°C threshold in this range architectural feature acting effectively as a greenhouse
and, consequently, they may be considered to have more incorporated to the building volume and acting as heat
chances to overheat. collector. However inappropriate use can lead to
Another noticeable finding that has emerged unwanted heat gains into the main house and higher
from this work is the distribution of hours in the range temperatures, contributing further to overheating (in this
“between 23°C and 26 °C". This temperature range can specific case study, a temperature of up to 42°C was
be seen as at high risk of overheating, since registered in the living room), see fig. 3.
temperatures can quickly increase above the threshold.
This is due to the very nature of highly insulated homes,
where internal temperatures are responding rapidly to
heat gains. In this respect, houses UK52 and UK54
showed the fewest hours between 23°C and 26 °C: 52%
and 43% respectively. By contrast, house UK51 showed
77% and 74% of hours between 23°C and 26°C in
bedrooms 1 and 2 respectively.
When considering the mean, minimum and
maximum temperatures, house UK51 and UK52 have
been found by their occupants to be too warm in the Figure 6: images of sunspace (winter garden) in house UK55.
bedroom (often up to the point that the occupants had to
move to another room in order for them to sleep well). Other monitored studies have shown that overheating
The mean temperatures in those bedrooms were above occurs in similarly conceived homes in England, Wales
23°C. For this reason, one might consider that the and Scotland, where also predictive tools deployed have
CIBSE general indoor comfortable operative failed to identify overheating risks, and where also
temperature for bedrooms, as listed in CIBSE Guide A occupant behaviour has a high influence of
(2006) of 23°C might need careful consideration when uncomfortably high temperatures. Similarly, these
designing or assessing comfort and well-being in highly studies also report that there is a lack of external solar
insulated dwellings. As far as the maximum shading which is contributing to unnecessary heat gains
temperatures are concerned, house UK51 recorded the (Morgan et al., 2015; Ridley, Bere, Clarke, Schwartz, &
highest temperatures in the bedrooms. After surveying Farr, 2014; Tabatabaei Sameni et al., 2015).
this house (UK51) and taking into account the responses
from the occupants, it can be hypothesised that the CONCLUSION
design of the house (open stack) may well have led to In general terms, it seems evident that the current
higher temperatures due to the exacerbation of ‘new’ way of designing homes in UK has perhaps not
temperature increase, typical of highly insulated matured yet an understanding of how to innovate
dwellings and further contribution of the open stack in architecture in consideration of a -much needed- low
the top floor bedroom, as represented in figure 5. carbon design. In fact, this study has provided evidence
that uncomfortable temperatures were found in all the
houses under review. However, this has occurred with
different degree of severity and apparently for a variety
of reasons.
One of the factors that most impact on
overheating experiences appears to be the presence or
absence of natural ventilation. In fact, this study showed
that in the houses where natural ventilation is applied
consistently temperatures were reduced.
Also the study showed that the lack of solar
control in general leads to excessive heat gains and
quick response in temperature increase. This suggests
that passive devices, such the sunspace (winter garden),
should be considered a double-edged sword: if
Figure 5: Representation of the stack effect through the improperly used or unmanaged, in highly insulated
stairwell in house UK51. buildings they may contribute an interesting space and
useful source of heat outside warm periods, but also
Another interesting design-related risk choice is exacerbate overheating if improperly used.
provided by UK55. In this case the highest temperatures The knowledge gained from these monitored case
were recorded in the sunspace/winter garden (see fig. 6). studies can be summarised in the following lessons
This incorporated sunspace (winter garden) is an learnt:
 PLEA 2016 Los Angeles - 32th International Conference on Passive and Low Energy Architecture.
Cities, Buildings, People: Towards Regenerative Environments

1. Mechanical ventilation in dwellings is for fresh air, https://www.gov.uk/government/uploads/system/uploads

not for summer cooling; both occupants as well as /attachment_data/file/7604/2185850.pdf
designers need to understand this and the need to 8. DCLG. The Building Regulations 2010. Conservation of
use additional natural ventilation in warm weather. fuel and power L1A. Conservation of fuel and power in
new dwellings 2013 Edition (2013). NBS.
2. Solar gains can cause severe overheating in highly 9. Dengel, A., & Swainson, M. (2012). Overheating in new
insulated homes even in the UK, where shading has homes. A review of the evidence (No. NF46). Milton
historically rarely been needed or used. Appropriate Keynes: NHBC Foundation.
window design and shading is therefore required. 10. HM Government. (2011). The Carbon Plan: Delivering
3. There is a need for greater use of detailed our low carbon future. London: Crown Copyright.
simulation at the design stage, particularly Retrieved from
regarding solar gains and ventilation, could be http://www.decc.gov.uk/en/content/cms/tackling/carbon_
employed to improve designs; simpler tools may plan/carbon_plan.aspx#
not inadequate for this sort of design, particularly in 11. Morgan, C., Foster, J., Sharpe, T., & Poston, A. (2015).
Overheating in Scotland: Lessons From 26 Monitored
the case of less experienced designers. Low Energy Homes. CISBAT 2015 International
Conference “Future Buildings and Districts -
Sustainability from Nano to Urban Scale”., 167–172.
ACKNOWLEDGEMENTS Retrieved from http://radar.gsa.ac.uk/3719/
The authors would like to acknowledge the 12. Nicholls, R. (2008). The Green Building Bible: The Low
sponsorship of the Global Innovation Initiative (GII) Energy Design Technical Reference. Green Building
project (2014-2016) “Reducing global energy use in Press. Retrieved from
http://books.google.co.uk/books?id=0Lv-oAEACAAJ
buildings while improving occupant comfort and well- 13. Ridley, I., Bere, J., Clarke, A., Schwartz, Y., & Farr, A.
being: reversing the growing trend toward energy- (2014). The side by side in use monitored performance
intensive air-conditioning”, funded by the British of two passive and low carbon Welsh houses. Energy
Council as part of an international partnership between and Buildings, 82, 13–26.
BERG-Loughborough University (UK), CBE-UC http://doi.org/10.1016/j.enbuild.2014.06.038
Berkeley (USA), CARBSE-CEPT University (India), 14. Tabatabaei Sameni, S. M., Gaterell, M., Montazami, A.,
IESD-De Montfort University (UK). Also, authors & Ahmed, A. (2015). Overheating investigation in UK
would like to express their gratitude to Dr Lynda Webb social housing flats built to the Passivhaus standard.
Building and Environment, 92, 222–235.
for her useful comments on an earlier draft of this paper.
http://doi.org/10.1016/j.buildenv.2015.03.030
In addition, the authors would like to express 15. Taylor, M. (2014). Preventing Overheating. London:
their most sincere gratitude to the very patient and Good Homes Alliance. Retrieved from
helpful volunteers of the houses surveyed. http://gha.pht.surefirehosting.co.uk/downloads/pages/RE
PORT GHA Preventing Overheating - FINAL
140217.pdf
REFERENCES
1. Beizaee, A., Lomas, K. J., & Firth, S. K. (2013).
National survey of summertime temperatures and
overheating risk in English homes. Building and
Environment, 65, 1–17.
2. CIBSE. (2006). Guide A: Environmental Design.
London: Chartered Institution of Building Services
Engineers.
3. CIBSE. (2013). The limits of thermal comfort : avoiding
overheating in European buildings. London: Chartered
Institution of Building Services Engineers.
4. CIBSE. (2015). Guide A: Environmental Design.
London: Chartered Institution of Building Services
Engineers. http://doi.org/10.1016/B978-0-240-81224-
3.00016-9
5. Crown. Climate Change Act 2008 (2008). UK: The
Stationery Office Limited. Retrieved from
http://www.legislation.gov.uk/ukpga/2008/27/pdfs/ukpg
a_20080027_en.pdf
6. Davies, M., & Oreszczyn, T. (2012). The unintended
consequences of decarbonising the built environment: A
UK case study. Energy and Buildings, 46, 80–85.
http://doi.org/10.1016/j.enbuild.2011.10.043
7. DCLG. (2012). Investigation into Overheating in
Homes. London: Crown Copyright. Retrieved from

View publication stats