You are on page 1of 67

Ventilation and Indoor Air Quality in Part F 2006 Homes

BD 2702
www.communities.gov.uk
community, opportunity, prosperity
Ventilation and Indoor Air Quality in Part F 2006 Homes
BD 2702
Simon McKay and David Ross
AECOM Ltd
Ian Mawditt and Stuart Kirk
Building Services Ltd
March 2010
Department for Communities and Local Government
Building Sciences Ltd
Department for Communities and Local Government
Eland House
Bressenden Place
London
SW1E 5DU
Telephone: 0303 444 0000
Website: www.communities.gov.uk
© Queen’s Printer and Controller of Her Majesty’s Stationery Office, 2010
Copyright in the typographical arrangement rests with the Crown.
This publication, excluding logos, may be reproduced free of charge in any format or medium for
research, private study or for internal circulation within an organisation. This is subject to it being
reproduced accurately and not used in a misleading context. The material must be acknowledged as
Crown copyright and the title of the publication specified.
Any other use of the contents of this publication would require a copyright licence.
Please apply for a Click-Use Licence for core material at www.opsi.gov.uk/click-use/system/online/
pLogin.asp, or by writing to the Office of Public Sector Information, Information Policy Team, Kew,
Richmond, Surrey TW9 4DU
e-mail: licensing@opsi.gov.uk
If you require this publication in an alternative format please email
alternativeformats@communities.gsi.gov.uk
Communities and Local Government Publications
Tel: 030 0123 1124
Fax: 030 0123 1125
Email: product@communities.gsi.gov.uk
Online via the Communities and Local Government website: www.communities.gov.uk
March 2010
Product Code: 09BD06323
ISBN: 978 1 4098 2375 9
The findings and recommendations in this report are those of the authors and do not necessarily
represent the views of the Department for Communities and Local Government.
Contents
Executive summary 5
Chapter 1 9
Introduction
Chapter 2 10
Background
2.1 Introduction 10
2.2 Primary aim 10
2.3 Secondary aims 11
Chapter 3 12
Study design
3.1 Dwelling selection and recruitment 12
3.2 Sample size 13
3.3 Monitoring period 13
3.4 Ventilation conditions 14
3.5 Monitoring 15
Chapter 4 18
Methodology
4.1 Air permeability 18
4.2 Ventilation capacity 18
4.3 Air exchange rates 19
4.4 TVOCs 19
4.5 HCHO 20
4.6 Nitrogen dioxide 20
4.7 Temperature and relative humidity 20
Chapter 5 21
Pilot study
5.1 Results and lessons learnt 21
Chapter 6 22
Results
6.1 Summary of dwellings monitored 22
6.2 Measurements of air permeability 22
6.3 Natural ventilation opening areas 25
6.4 Whole house air exchange rates 28
6.5 Intermittent extract flow rates 31
6.6 Relative humidity 34
6.7 Nitrogen dioxide 37
6.8 TVOCs 38
6.9 Formaldehyde 39
Chapter 7 40
Discussion
7.1 Are intermittent extract flow rates sufficient? 40
7.2 Are trickle ventilator areas sufficient? 42
7.3 Are dwellings Part F compliant? 51
7.4 Are dwellings Part L compliant? 51
7.5 Should we change air pressure testing for Part L compliance? 52
7.6 Should we undertake air pressure testing for Part F compliance? 52
Chapter 8 53
Conclusions and recommendations
Appendix A 56
Detailed results
Executive summary
|
5
Executive summary
Introduction
1. This study was commissioned by Communities and Local Government to inform
and provide evidence for the amendments to the Part F regulations and guidance
documents that are due to come into force in October 2010.
2. The study has assessed whether the guidance in the 2006 edition of Approved
Document F (ADF) is effective at providing adequate ventilation and good indoor air
quality in new dwellings, thereby minimising the risks to health of the occupants.
It has also investigated how well dwellings comply with Part F 2006 standards. The
focus has been on more airtight dwellings which are naturally ventilated.
3. This report presents the results from a sample of 22 occupied homes built to Parts L and
F 2006 standards. The project involved carrying out measurements of airtightness,
whole house air exchange rates, mechanical extract flow rates, relative humidity
(RH) levels and other indoor pollutant concentrations. All homes were occupied
with, where possible, the ventilation system set up at full capacity to test whether
the ventilation guidance in ADF is adequate. Diary records of occupant activities and
questionnaires on the dwellings and their indoor environments were also collected.
4. The results have been analysed to assess the significance of dwelling characteristics,
ventilation factors and indoor sources in determining concentrations of indoor
pollutants. They have also been analysed to determine whether the design
recommendations in ADF 2006 are being met within the dwellings.
Approved Document F 2006
5. ADF 2006 provides guidance for new dwellings built to an air permeability down to
3 to 4 m
3
.h
-1
.m
-2
at 50 Pa. ADF accounts for the contribution made by air infiltration
when determining the purpose-provided ventilation necessary to provide good indoor
air quality. It suggests that additional ventilation may be required for more airtight
homes where there is less air infiltration, but no additional guidance is provided.
6. Since ADF 2006 was published, new evidence has emerged to suggest that there has
been a significant improvement in the airtightness of new dwellings. Approximately
5% of around 3000 new dwellings that have been pressure tested achieve results
better than 3 m
3
.h
-1
.m
-2
and 30% achieve better than 5 m
3
.h
-1
.m
-2
.
6
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Aims
7. The primary aim of this study was to establish whether additional ventilation is
required in more airtight dwellings with an air permeability equal to or better than
5 m
3
.h
-1
.m
-2
. The study focused on ADF ventilation system 1: “Background ventilators
with intermittent extract fans” because:
º This venlilalion syslem lype is lhe one mosl commonly used in new dwellings.
º Crealer concern has been expressed by slakeholders on lhe use ol nalural
ventilation in airtight properties, and in particular the reliance on natural
driving forces (wind and temperature stack effects) to provide the background
ventilation through trickle ventilation.
º Civen lhe sample size and lhe dillcully ol lnding dwellings wilh an air
permeability equal to or better than 5 m
3
.h
-1
.m
-2
, the analysis would be more
meaningful if it focused on just one ventilation system type.
8. There were a number of secondary aims:
º Compliance with Part F: We wished to investigate whether the installed
ventilation systems complied with Part F guidance. There is both objective
and anecdotal evidence to suggest that ventilation systems are designed to
comply with ADF 2006 standards but do not achieve this on construction due to
inadequate installation and inspection.
º Air pressure testing for Part L: For Part L compliance, air pressure testing is
undertaken to BS EN 13829: Method A (all mechanical ventilation systems
sealed, all natural ventilation openings closed). The proposed amendments
to ADL1A would change the test method to BS EN 13829: Method B (all
mechanical ventilation systems sealed, all natural ventilation openings closed
and sealed). We wished to investigate the impact and need for this change.
º Air pressure testing for Part F: We wished to investigate if an air pressure test
with all natural ventilation systems fully open would provide an estimate of the
equivalent area of the installed background (trickle) ventilation.
Results
9. The key results from this study are:
º We have monilored 22 dwellings buill lo Parls l and L 2006 slandards, wilh
trickle ventilation and intermittent extract. Ten of the 22 dwellings tested (46%
of total sample including pilot homes, 50% of main sample) achieved an air
permeability of less than 5 m
3
.h
-1
.m
-2
.
Executive summary
|
7
º Nearly lhree-quarlers ol lhe dwellings (72° ol sample) did nol have sullcienl
trickle ventilation installed to meet ADF 2006 guidance. In the worst case, only
just over half (57%) of the recommended trickle ventilation was installed.
º 52° ol lhe inlernal doors lailed lo achieve lhe !0 mm gap under lhe doors
recommended in ADF 2006.
º Less lhan hall ol lhe kilchen or balhroom lans achieved lhe low rales
recommended in ADF 2006. Furthermore, none of the fans located away from
the cooker achieved the recommended flow rates in the kitchen.
º All relalive humidily levels monilored over a week-long period were wilhin
the new daily and weekly relative humidity guideline levels proposed in the
consultation version of ADF 2010. However, making reasonable assumptions, it
would be expected that approximately four homes would have monthly relative
humidity levels that would exceed the newly proposed monthly relative humidity
guideline level.
º lour homes had levels ol nilrogen dioxide in lhe kilchen lhal exceeded lhal
recommended in ADF 2006. A key issue in these cases was insufficient installed
flow where fans are located away from the cooker.
º All homes had acceplable levels ol lormaldehyde.
º Over hall ol lhe homes wilh TVOC measuremenls had levels exceeding lhal
recommended in ADF 2006. It is important to note that in developing ADF 2006,
TVOCs were chosen to represent organic compounds. A review of indoor air
quality results suggested that if the TVOC guideline level of 300 μg.m
-3
was met,
then individual chemical health-based guideline levels (eg benzene, toluene,
formaldehyde) should also be met.
Conclusions
10. The key conclusions are:
º ll is dillcull lo make a conclusive judgemenl as lo whelher or nol lhe currenl
recommended natural ventilation provisions in ADF 2006 are sufficient for
airtight homes. The study is relatively small and the conclusions should only be
treated as indicative. Furthermore, in all cases, the capacity of the ventilation
system did not meet that recommended in ADF 2006.
º The inlermillenl exlracl rales appear lo be sullcienl. ll is imporlanl lo ensure
that they are installed correctly to provide the capacity recommended in ADF
2006. There is no evidence from this study that with correct installation, the
extract rates should be increased for more airtight dwellings.
º The lrickle venlilalor sizes appear lo be insullcienl. Lven allowing lor inslallalion
issues for the ventilation system as a whole, at least a significant minority of
TVOC levels would be expected to exceed the recommended guideline level,
8
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
particularly for the more airtight homes. VOCs are produced by building products
and the activities of the occupants such as smoking, use of personal hygiene
products and interior decorating. It may seem pedantic to change the design
of a ventilation system just to better control one pollutant, but the long term
effects on health due to exposure to VOCs are not well documented, and it
seems prudent to err on the side of caution. Furthermore, the TVOC criterion
was selected as a sensitive marker for individual organic chemical compounds (ie
if TVOC levels are below the criterion, previous research suggest that individual
organic chemical compounds would also be below recognised indoor or outdoor
health-based levels for these pollutants). There are also other hazards to health
that are affected by ventilation, such as house dust mites, which are associated
with respiratory illnesses including asthma. We do not currently attempt to
control these under Part F because their breeding success is influenced by
heating and hygiene practice as much as ventilation. Improving ventilation would
be step in the right direction to limit this risk.
º 8eller guidance needs lo be provided lor lhe nalural venlilalion ol lals. While
the flats had two façades, and were ventilated as ‘multi-sided façades’ according
to ADF 2006, they were effectively single façades as the second façade was
limited. The ventilation and indoor air quality results suggest that these homes
were under-ventilated.
º ll is imporlanl lo nole lhal in praclice, occupanls do nol use lheir venlilalion
system to full capacity (trickle vents are not always open and fans are not always
used). Hence, the actual pollutant levels in dwellings will likely be higher than
those recorded in this study.
Recommendations
11. The key recommendations are:
º ll is imporlanl lhal lhe venlilalion syslem should be inslalled correclly and
inspected to provide the ventilation capacity as designed.
º lurlher evidence needs lo be oblained lo subslanliale lhe indicalors lrom lhis
study. However, there are sufficient grounds to support the proposal to increase
trickle ventilators in dwellings having an air permeability equal to or tighter than
4 m
3
.h
-1
.m
-2
. This is based on the levels of TVOCs observed, and the implications
for individual organic compounds. This could also have benefits in controlling
other hazards to health, but this is difficult to quantify.
º This was a relalively small sludy. A larger sludy is necessary, building on lhe
findings of this study, to better determine indoor air quality levels achieved in
airtight dwellings in which the ventilation system is installed correctly.
º Civen lhe relalively high level ol TVOCs, grealer consideralion should be given lo
reducing source strength by product controls.
Chapter 1 Introduction
|
9
Chapter 1
Introduction
This study was commissioned by Communities and Local Government (CLG)
to assess whether the guidance in the 2006 revision of Approved Document F
(ADF) is effective at providing adequate ventilation and good indoor air quality
in new dwellings, thereby minimising the risks to health of the occupants. A key
secondary aim was to investigate how well dwellings complied with Part F 2006.
The study particularly focused on more airtight dwellings which are naturally
ventilated. It is intended that the results from this study will help inform the 2010
amendments to Parts F and L of the Building Regulations.
This report presents the results from a sample of 22 occupied homes built to Parts
L and F 2006. The project involved carrying out measurements of airtightness,
whole house air exchange rates, mechanical extract flow rates, relative humidity
(RH) levels and other indoor pollutant concentrations. All homes were occupied
with, where possible, the ventilation system set up at full capacity to test whether
the ventilation guidance in ADF is adequate. Diary records of occupant activities
and questionnaires on the dwellings and their indoor environments were also
collected.
Analysis of the results was used to assess the significance of dwelling
characteristics, ventilation factors and indoor sources in determining
concentrations of indoor pollutants. The results were also analysed to determine
whether the design recommendations in ADF 2006 were being met within the
dwellings.
10
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Chapter 2
Background
2.1 Introduction
This study was undertaken to provide evidence for the proposed amendments
to Part F 2010. In particular, it reviews the necessity of increasing ventilation
provisions for more airtight dwellings. It also addresses other issues such as
compliance with Part F.
2.2 Primary aim
ADF 2006 provides guidance for new dwellings built down to an air permeability
of 3–4 m
3
.h
-1
.m
-2
at 50 Pa. It accounts for the contribution of air infiltration
in determining the purpose-provided ventilation necessary. It suggests that
additional ventilation provisions may be required for more airtight homes, due
to less air infiltration. As it was expected that in the near future relatively few
dwellings would approach or be tighter than this level of air permeability, no
additional guidance was provided for these dwellings at that time. Approved
Documents are intended to provide guidance for some of the more common
building situations.
Since ADF 2006 was developed, new evidence suggests that there has been a
significant improvement in the airtightness of new dwellings. Approximately
5% of new dwellings pressure tested achieve results better than 3 m
3
.h
-1
.m
-2
and 30% achieve better than 5 m
3
.h
-1
.m
-2
. This is based on a total sample of
approximately 3000 new dwellings and arises from two sources which agree with
one another: a confidential industry analysis and Building Sciences Ltd (BSL) air
pressure testing results.
Hence additional guidance is proposed in the consultation version of ADF 2010
for the more airtight homes. In the main, the ventilation system specifications
are not amended for dwellings designed to an air permeability leakier than
5 m
3
.h
-1
.m
-2
. However, increased ventilation provisions are recommended for
dwellings designed to be tighter than this.
The primary aim of this study is to provide evidence to determine whether
additional ventilation provisions are required for those more airtight dwellings,
Chapter 2 Background
|
11
with air permeability equal to or better than 5 m
3
.h
-1
.m
-2
. Furthermore, it
particularly focuses on ADF ventilation system 1: background ventilators with
intermittent extract fans. There are three key reasons for this focus:
º This venlilalion syslem lype is lhe one mosl commonly used in new dwellings.
º Crealer concern has been expressed by slakeholders on lhe use ol nalural
ventilation in airtight properties. In particular, the reliance on natural driving
forces (wind and temperature stack effects) to provide the background
ventilation through trickle ventilation.
º Civen lhe sample size, and praclical dillcullies in achieving only dwellings
with an air permeability equal to or better than 5 m
3
.h
-1
.m
-2
as described
later, better analysis could be undertaken by focusing on just one ventilation
system type – with the expectation that inferences could be made to other
system types.
2.3 Secondary aims
There were a number of secondary aims to this study.
Review compliance with Part F: We wished to investigate whether the
installed ventilation systems complied with Part F. There was a mix of objective
and anecdotal evidence to suggest that the ventilation systems were designed to
ADF 2006 but did not achieve this on construction due to inadequate installation
and inspection.
Air pressure testing for Part L: For Part L compliance, air pressure testing is
undertaken as per BS EN 13829: Method A (all mechanical ventilation systems
sealed, all natural ventilation openings closed). The consultation version of
ADL1A proposes changing the test method to BS EN 13829: Method B (all
mechanical ventilation systems sealed, all natural ventilation openings closed and
sealed). We wished to investigate the impact and need for this change.
Air pressure testing for Part F: We wished to investigate the use of a further
air pressure test for Part F. This would be undertaken similarly to the testing above
but with all natural ventilation systems fully open. The proposal was that by
comparing these results with those of Method B (with natural ventilation systems
fully sealed), it could provide an estimate of the equivalent area of background
(trickle) ventilation installed in dwellings to confirm installed capacity met design
intentions.
12
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Chapter 3
Study design
3.1 Dwelling selection and recruitment
The initial selection criteria for dwellings were as follows.
º Conslrucled lo Parls l and L 2006
º Mixlure ol lypes (lal, delached house, lerraced house, bungalow, elc)
º Air permeabilily equal lo or beller lhan 5 m
3
.h
-1
.m
-2
º Occupied al leasl six monlhs prior lo sampling lo reduce lhe inilial impacl ol
high indoor pollutant emissions from, for example, the building drying-out,
new construction materials and furnishings, and painting and decorating. No
specific occupancy levels were selected – although we looked to obtain the
sample from both private and social housing to provide a mix.
º Wilhin a reasonable dislance (!50 miles) ol Oxlord where 8uilding Sciences
Limited, who were carrying out the testing, are located.
These initial criteria proved challenging for three principal reasons.
º While house-builders were inleresled in lhis sludy, lhere were limils lo lhe help
they could give in finding these homes given the current economic climate.
º Pelalively lew dwellings had been conslrucled lo 2006 8uilding Pegulalions
and lived in for six months. Many dwellings constructed since 2006 had
planning permission to pre-2006 Building Regulations.
º Where dwellings were idenliled lhal mel lhe above crileria, il lhe inilial
contact was with the householder, in a majority of the cases they did not have
details of the airtightness of their dwelling to determine if it was equal to or
better than 5 m
3
.h
-1
.m
-2
.
As a result, we modified the criteria to monitor any dwelling constructed to
Part F/L 2006 (air permeability should be better than 10 m
3
.h
-1
.m
-2
). This in itself
is a valuable study as we are assessing the performance of newly introduced
regulations.
To meet the primary aim of the study, we would aim to gain data on the
performance in the most airtight homes through:
Chapter 3 Study design
|
13
º biasing lhe sample where possible lo lhose more airlighl dwellings wilh an air
permeability better than or equal to 5 m
3
.h
-1
.m
-2
º analysing bolh lhe sub-sel ol more airlighl dwellings as well as invesligaling
the trend in performance from leakier to more airtight dwellings.
We decided to focus only on ADF ventilation system 1 (trickle ventilation with
intermittent extract). This should allow us to better determine the trend line
in performance as airtightness varies (ie removes variation due to different
ventilation system types). Additional reasons for focusing on this ventilation
system were discussed in Section 2.
The sample was obtained through a combination of the following routes:
º an invilalion lo ALCOM slall via lhe inlranel
º 8uilding Sciences air pressure lesling dalabase
º induslry conlacls
º mailing households on new developmenls
º lealel drop on new developmenls.
A disturbance payment of £100 was paid to the householder to take part in the
study. Information about the nature of the testing was provided in advance.
3.2 Sample size
An initial pilot study of two dwellings was undertaken. The aim of the pilot study
was to assess the protocol developed for this project in terms of practicality,
reliability and variability, to confirm the techniques for measurements and to
check the suitability of the household questionnaires and diaries that had been
developed for the study.
The sample size for the main study was 20 dwellings with the intention of at
least half having an air permeability of better than 5 m
3
.h
-1
.m
-2
. While the total
sample number was relatively small, it was considered that it should give a good
indication if issues of poor indoor air quality or poor compliance with Part F were
common.
3.3 Monitoring period
The intention was to undertake the main sampling in March and April 2009.
14
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
º Nalural venlilalion is dependenl on wind and slack driving lorces. During lhis
period, the difference between internal and external temperatures would be
reduced and thus the stack driving forces lower.
º lnlernal PH levels would likely be higher as lhe exlernal moislure levels are
high and there is less warming of the outside air (and reduction in RH) as
the air enters the home. (It is noted that the actual risk of condensation
should also take account of the internal surface temperatures, which would
be lowest during the coldest winter periods. However, this methodology
provides the reasonably worst case internal RH levels.)
º ll would slill be during lhe healing season and so lhe dwelling would be lairly
‘closed-up’, i.e. relatively little use of windows, etc.
In practice, this aim was broadly met with the main study taking place between
March and mid-May. The initial pilot study took place in January/February 2009.
Each dwelling was monitored for one week. It was thought sufficient to provide
the information required for that dwelling. Furthermore, as highlighted below,
the study made requirements of the occupants and based on experience it was
considered that the necessary occupant behaviour may not have been reliably
met over longer periods.
3.4 Ventilation conditions
To assess whether ADF 2006 recommendations were adequate, it was necessary
for the ventilation system to be used to its full capacity. Hence the following steps
were undertaken:
º all lrickle venlilalors (or olher venlilalion inlels) were lully opened and lhe
occupants asked to keep them open
º occupanls were asked lo use balhroom/WC mechanical exlracls al lull
capacity during all bathing occasions
º occupanls were asked lo use lheir kilchen exlracl on ils highesl selling
during cooking times
º occupanls were asked nol lo open windows.
A daily diary was given to the occupants to complete during the monitoring
period to confirm that the above instructions were followed and, if not, why they
needed to make changes.
Chapter 3 Study design
|
15
3.5 Monitoring
3.5.1 Introduction
The strategy of the study was based on previous research experience of
measuring airtightness, ventilation and indoor air quality in homes and used as
far as possible well proven, validated techniques and methods of investigation
in order to provide reliable, quality assured data. A summary of the parameters
measured in this study is provided in Table 1. The parameters selected are further
discussed below and the next section provides more detail of the individual
measurement techniques.
3.5.2 Airtightness
The following airtightness tests were carried out on the first visit to each dwelling:
º 8S LN !3829. Melhod A (all mechanical venlilalion syslems sealed, all nalural
ventilation openings closed)
º 8S LN !3829. Melhod 8 (all mechanical venlilalion syslems sealed, all nalural
ventilation openings closed and sealed)
º 8S LN 832 wilh all mechanical venlilalion syslems sealed, all nalural
ventilation openings opened.
Comparison of the first two tests will allow an evaluation of the advantage of
including Method B instead of Method A
1
as the means for compliance testing for
dwellings.
Comparison of the latter two tests provides a means of estimating the equivalent
area of trickle ventilation in the dwelling to compare against what appears to
have been installed.
3.5.3 Ventilation
The perfluorocarbon (PFT) technique was used to provide an estimate of the
ventilation rate in the dwelling over the one week sampling period.
3.5.4 Extract flow rates
A rotating vane anemometer with an aircone hood attachment was used to
determine the air flow rate of each extract fan and cooker hood in the dwellings
at its highest setting.
1
Technical Standard 1 (TS1), published by the Air Tightness Testing and Measurement Association (ATTMA), currently refers to BS EN
13829 Method B as the compliance testing method. However, Method B is specifically amended in TS1 to remove the requirement to
temporarily seal trickle ventilators. Hence, for the purpose of this report, the method referenced for Building Regulations compliance is
Method A.
16
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
3.5.5 Indoor air quality
The following indoor pollutants were monitored:
º Pelalive humidily (PH)
º Nilrogen dioxide (NO
2
)
º Tolal volalile organic compounds (TVOCs) and some individual VOCs
º lormaldehyde (HCHO).
All of these pollutants, with the exception of formaldehyde, are covered by
performance criteria in ADF Appendix A. Formaldehyde was also measured as
there have been suggestions to introduce formaldehyde into the Appendix A
performance standards.
3.5.6 Questionnaires/diaries
Each occupant who participated in the study answered a questionnaire
concerning his/her home, their normal ventilation behaviour and the activities
in the home during the sampling period. In addition, the occupant completed a
diary regarding gas cooking activity, use of extract fans and window opening.
During the initial visit to the property, the research team recorded:
º lhe sizes ol lrickle venlilalors in each room
º whelher lrickle venlilalors were open or closed
º lhe size ol lhe inlernal door underculs (belween lhe bollom ol lhe door and
the top of the floor finish)
º lhe selling lound on lhe cooker hood or exlracl lan in lhe kilchen
º lhe posilion ol lhe isolalor swilches (ie on or oll) lor balhroom, en-suile and
WCs
º over-run limes lor all lans.
This was undertaken to provide information about how the occupants would
normally operate their ventilation system, without being biased from later stages
of the study, after we had informed them of the appropriate utilisation of the
ventilation system and reasons for doing so.
Chapter 3 Study design
|
17
Table 1 Summary of the IAQ and ventilation monitoring during the pilot and
main study
Pilot study Main study
Sampling period 29 Jan to 5 Feb 2009 2 March to 15 May 2009
No. of homes 2 20
IAQ parameters
and locations
monitored
NO
2
(kitchen and outside)
TVOCs (living room, master
bedroom and outside)
Temperature and humidity
(living room, master
bedroom, and outside)
HCHO (living room, master
bedroom and outside)
NO
2
(kitchen and outside)
TVOCs (living room, master
bedroom and outside)
Temperature and humidity
(living room, master
bedroom, kitchen,
bathroom and outside)
HCHO (living room, master
bedroom and outside)
Ventilation rate
measurements
Air permeability
measurements (three
types)
PFT technique for whole
house ventilation
Internal door undercuts
Mechanical extract fan
flow rates (bathroom, en-
suites, WC and kitchen)
Air permeability
measurements (three
types)
PFT technique for whole
house ventilation
Trickle ventilator area
noted (not measured) in
each room
Internal door undercuts
Mechanical extract fan
flow rates (bathroom, en-
suites, WC and kitchen)
18
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Chapter 4
Methodology
4.1 Air permeability
The dwellings were air permeability tested using the following three
methodologies:
º As per 8S LN !3829. Melhod A. All mechanical venlilalion syslems sealed, all
natural ventilation openings closed.
º As per 8S LN !3829. Melhod 8. As above bul wilh nalural venlilalion
openings closed and sealed.
º As per 8S LN 832. As Melhod A bul wilh all nalural venlilalion inlels lully open.
Each of the above test methods was performed using pressurisation and
depressurisation procedures, ie six tests per dwelling, using UKAS calibrated
domestic standard ‘blower door’ fan and micromanometers and associated
equipment. The final air permeability results for each method are determined
as the average of the pressurisation and depressurisation results with an overall
uncertainty of less than ±10%.
4.2 Ventilation capacity
4.2.1 Natural
Approved Document F (ADF) recommends that the equivalent area is marked on
the trickle ventilator in any easily visible location where practical. Where available,
the equivalent areas were recorded. In the few cases where this information was
not visible, a geometric area measurement of the trickle ventilator was made. The
total equivalent area for each dwelling and associated dwelling floor areas were
referenced to the recommended equivalent areas published in Table 1.2a of ADF.
Door undercuts to internal doors were measured to confirm adequacy of cross
ventilation provision, ie 10 mm clear gap.
4.2.2 Mechanical
The volumetric flow through all intermittent extract fans and cooker hoods was
measured within each dwelling using a UKAS calibrated Airflow AV-2 rotating
vane anemometer with an aircone hood attachment to encompass the fan, or
Chapter 4 Methodology
|
19
ceiling/wall terminal. The readings were all taken with the fans and hoods at their
highest flow setting over a 30 second averaging time. The overall uncertainty
with this measurement technique for the flow rates measured is less than ±5%.
4.3 Air exchange rates
The PFT technique is a passive sampling technique for measuring ventilation rates
in buildings, which is described in ISO-standard 16000-8. In this technique, a
perfluorocarbon compound is passively emitted from small tracer sources, with a
known constant emission rate, and passively collected on adsorption tubes. The
amount of tracer adsorbed depends mainly on the emission rate from the source
tubes and the dilution of it by ventilation air. The PFT technique is based on the
fact that the average air infiltration rate is approximately equal to the reciprocal of
the time averaged indoor tracer gas concentration and, using this technique, the
time averaged indoor tracer gas concentration is determined.
PFT sources were set out in all rooms in each house, including hallways and
stairwells, but not in small or wet rooms (e.g. store cupboards and bathrooms).
The total number of sources in each room (source strength) was arranged to give
an approximately uniform source rate per unit volume throughout the house. The
samplers were placed in the living rooms, kitchen and all bedrooms.
The PFT sources and samplers were obtained from PentIAQ, Sweden. Following
the sampling, they were then returned to PentIAQ for analysis. The results for the
effective air exchange rate for the houses are derived from the emission rate per
volume (room) and the local mean age of air.
The overall uncertainty of this technique is a combination of the source emission
rate, the sampling rate, site limitations and number of sources/samplers, and the
laboratory analysis, which combined, provides an accuracy of ± 10%.
4.4 TVOCs
Passive automatic thermal desorption (ATD) tubes, pre-conditioned with Tenax
®
TA at the analytical laboratory (Scientifics), were installed in each house for a
period of seven days. Three ATD tubes were installed at each house: the master
bedroom, living room and garden. After the sampling period, the tubes were
analysed at the laboratory by automatic thermal desorption–gas chromatography
(ATD–GC). Chemical compounds were then identified via a mass spectrometric
(MS) detection library. The concentration of total volatile organic compounds
(TVOCs) was determined by the total ion current response of the individual
compounds, quantified using the calibrated response factor of toluene.
20
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
4.5 HCHO
Passive formaldehyde (HCHO) badges were installed for three days consecutive
sampling using SKC UNME00 samplers using 2,4dinitrophenyl-hydrazone
(DNPH) as the adsorbent packing material. The HCHO badges were supplied
and analysed by the analytical laboratory (Scientifics), and were installed in the
master bedroom, living room and garden of each house. Desorption and analysis
of the diffuse samplers was performed using a diffuse high Performance liquid
chromatography (HPLC) using UV absorption between 230 and 370 nm. Limit
of detection (LOD) was 1.0 μg.m
-3
with an accuracy of ±15%. The concentration
of HCHO in the air was determined by subtraction of HCHO adsorbed onto the
blank (correction) tape from that adsorbed onto the exposed sample tape.
4.6 Nitrogen dioxide
The average mean concentration of NO
2
was measured using passive diffusion
(Palmes) tubes over the 7-day sampling period. The Palmes tubes were supplied
and analysed by the analytical laboratory (Scientifics) and were installed in the
kitchen and garden of each property, at least 1 metre from vertical surfaces. This
method relies on the transfer of NO
2
by diffusion to a collector at one end of
the tube, which contains a mesh treated with 50% triethanolamine (TEA) and
50% acetone that adsorbs NO
2
. At the end of the sampling period, the mean
NO
2
concentration was quantified using a segmented flow autoanalyser with
ultraviolet detection. The reported accuracy is ±10%.
4.7 Temperature and relative humidity
Hygrothermal conditions were recorded using combined temperature and
relative humidity USB data loggers located in the master bedroom, living room,
kitchen, main bathroom and garden of each property. The loggers were set
to record at 5-minute intervals during the 7-day monitoring period and were
generally placed between 1000–1500 mm above floor level to minimise
variations induced by thermal stratification. At the end of the monitoring, the
data were downloaded for later analysis. The typical accuracy of the loggers used
was ±0.3% and ±2% for temperature and relative humidity, respectively.
Chapter 5 Pilot study
|
21
Chapter 5
Pilot study
5.1 Results and lessons learnt
The aim of the pilot study was to assess the protocol, measurement techniques
and questionnaires and diaries. Overall, everything worked as intended and
as a result we have included the pilot study results with the main study results
(see Section 7). RH was not measured in the kitchen and the bathroom in the
pilot homes but was in the main study. In addition, the pilot study only recorded
the number of trickle vents whereas in the main study we also recorded the
equivalent area.
The main lesson learnt was the difficulty in obtaining the sample group (discussed
previously in Sections 2 and 3). This resulted in modifications to the study design,
in particular the criteria for recruitment.
22
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Chapter 6
Results
6.1 Summary of dwellings monitored
The sample of 22 homes (including the pilot study), comprised the following
dwelling types all built to Parts F and L 2006.
º lve lals
º six lerraced houses (lhree mid and lhree end)
º lve semi-delached houses
º six delached houses.
6.2 Measurements of air permeability
6.2.1 Introduction
This section provides a summary of the results of the air permeability tests
undertaken at all 22 homes. The individual test results for each dwelling are
presented in Table A.1 of the Appendix.
6.2.2 Air pressure testing with trickle vents closed (Method A)
As referred to in Section 4.1, air pressure test Method A is the current test method
used for Part L compliance purposes. All mechanical ventilation systems are
sealed and all natural ventilation openings are closed but not sealed.
Figure 1 shows the distribution of the air permeability for each dwelling. Ten of
the 22 dwellings tested (46% of total sample including pilot homes; 50% of
main sample) achieved an air permeability of less than 5 m
3
.h
-1
.m
-2
. Two homes
exceeded the maximum level of 10 m
3
.h
-1
.m
-2
required by ADL 2006 for dwellings
sampled. This is discussed further in Section 7.
Figure 2 shows how the air permeability varies by dwelling type (P1–P2 are the
pilot homes and H1–H20 are the main study homes). Overall, the flats were the
most airtight dwelling type in this sample – all achieving an air permeability of less
than 5 m
3
.h
-1
.m
-2
, with a mean of 3.7 m
3
.h
-1
.m
-2
. The mean air permeability of the
remainder of the dwellings (houses) was 6.9 m
3
.h
-1
.m
-2
. Each dwelling type had
at least one dwelling with an air permeability of better than 5 m
3
.h
-1
.m
-2
. Note
that while we looked to bias this sample towards more airtight dwellings, the
Chapter 6 Results
|
23
overall mean value of 6.2 m
3
.h
-1
.m
-2
is similar to the mean value of approximately
6.5 m
3
.h
-1
.m
-2
obtained from a sample of approximately 3000 new dwellings
combined from two sources: a confidential industry analysis and Building
Sciences Ltd (BSL) unpublished air pressure testing results (from their air pressure
testing services).
Figure 1 Air permeability distribution (Method A)
Figure 2 Air permeability distribution by dwelling type
24
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
6.2.3 Air pressure testing with trickle vents sealed (Method B)
As referred to in Section 4.1, air pressure test Method B was similar to Method A
with the addition of all natural ventilation openings (in this case trickle ventilation)
being sealed instead of simply closed. This test method is proposed in the
consultation version of Part L 2010 to replace Method A.
The results show a small reduction in air permeability with trickle ventilation
sealed. Figure 3 shows this reduction in air permeability. The mean reduction was
3.3% or 0.2 m
3
.h
-1
.m
-2
@ 50 Pa.
The relative merits of Methods A and B are discussed in Section 7.
Figure 3 Reduction in air permeability with vents sealed
6.2.4 Air pressure testing with trickle vents open
As referred to in Section 4.1, these tests were similar to Methods A and B. The
exception is that all natural ventilation openings (in this case trickle ventilation) are
fully open.
The results show a smaller than expected difference compared to the results with
trickle vents sealed (Method B). The results are shown in Figure 4. They show an
increase of only 2.6 m
3
.h
-1
.m
-2
@ 50 Pa on average on fully opening the trickle
ventilators. In the development of ADF 2006, and using the 1/20th rule-of-thumb
conversion between air permeability and infiltration, we would have expected
a value perhaps twice this value. Causes for these low readings are discussed
further in this and the next section.
Chapter 6 Results
|
25
Figure 4 Increase in air permeability with trickle ventilators open
6.3 Natural ventilation opening areas
6.3.1 Introduction
This section provides a summary of the results of the ventilation provisions
installed. For results of individual dwellings see Table A.2 in the Appendix.
6.3.2 Recorded trickle ventilation equivalent areas
Figure 5 shows the provision of trickle ventilation in the dwellings compared to
that recommended in ADF (2006) Table 1.2a. Of the sample of 22 dwellings, data
were not recorded in the pilot study and only 18 of the 20 homes in the main study
had the equivalent area visible on the trickle ventilator and are reported here.
Nearly three-quarters of the dwellings (72% of sample) did not have sufficient
trickle ventilation installed to meet ADF (2006). At worst, just over half (57%)
of the recommended trickle ventilation was installed. In contrast, the maximum
provision of trickle ventilation was 162% of that required in a home with a high
window to façade ratio.
Further review of the data suggests that the provision of trickle ventilation in
many homes was based on one trickle ventilator per window installed rather
than reviewing the overall requirement of the dwelling. This may be insufficient,
depending on the size of trickle ventilator installed.
This under-provision of trickle ventilation will partially explain the lower than
expected readings for air pressure test Method C (see Section 6.2). However, this
is not expected to be the full explanation. The results from the air pressure tests
26
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
suggest that we are only installing 50% of the required trickle ventilator area on
average. The area of trickle ventilation recorded provides a possible explanation
for 6% (ie one-tenth) of this.
It is also worth noting that only 40% of the trickle ventilators were open when the
monitoring team arrived at the property (see Table A.3 in the appendix). For the
purpose of this study, the trickle ventilators were all left in the open position.
Figure 5 Installed trickle ventilation versus ADF Table 1.2a requirement
57%
6.3.3 Measured trickle ventilation equivalent areas
An alternative method was used to determine the equivalent area, using
Methods B and C pressure testing results. This was to assess whether it was
possible to measure simply on-site whether the equivalent area written on the
trickle ventilator area was accurate. Two sources of errors in the laboratory testing
were considered possible. This may provide a further explanation of the lower
than expected readings for air pressure test Method C (see Section 6.2).
º The lrickle venlilalor should be inslalled in lhe laboralory lesls in similar
surroundings to the actual installation and this may not be the case.
º The general accuracy ol lhe lesling procedure.
Three steps were undertaken in analysis:
Step 1: The total equivalent area at 1 Pa was determined for each dwelling using
Method B (trickle vents sealed) and Method C (trickle vents open), using the
following equation:
A=1000.Q
50
.(P
1
/P
50
)
n
.(1/C
d
).(r/2P
1
)
0.5
Chapter 6 Results
|
27
where:
A = the background ventilator equivalent area (mm
2
)
Q
50
= the air supply rate at 50 Pa (l/s)
C
d
= the discharge coefficient, taken as 0.61
r = air density (kg/m
2
), taken as 1.2
P
1
, P
50
= the pressure across the vent at 1 and 50 Pa
n = flow rate exponent for the air pressure test.
Step 2: For each of the two methods, the equivalent area was calculated using ‘n’
from pressurisation and ‘n’ from depressurisation fan tests and an average taken.
Step 3: The equivalent area from Method B was subtracted from Method C.
This should give the equivalent area for the trickle ventilators only at 1 Pa. The
equivalent area reported on a trickle ventilator is also the equivalent area at 1 Pa.
The results were generally smaller than the total equivalent areas recorded on the
trickle ventilators. However, it is considered that the results from this analysis are
not sufficiently reliable. The air pressure testing results were at 35 Pa and above.
Within the equation, the flow rate exponent (‘n’) is used to extrapolate down to
1 Pa. The evaluation of ‘n’ is not sufficiently accurate to do so in a robust manner.
It would be necessary to undertake fan flow rate tests at lower pressures to
improve this accuracy.
6.3.4 Door undercuts
Natural ventilation relies on air flow between rooms – cross ventilation across
the dwelling and/or stack ventilation up through the dwelling. When internal
doors are closed, there needs to be sufficient gap around the doors to maintain
adequate air flow. To achieve this, ADF (2006) recommends that doors have a
minimum undercut of 10 mm above the floor finish.
Figure 6 summarises the door undercut in the sampled dwellings. Of the 127
doors measured, 52% of all doors failed to achieve this 10 mm gap. Hence,
these rooms may well be under-ventilated when their door is closed as well as
the ventilation for the whole house being lower as the contribution of the trickle
ventilation in that room to the whole house ventilation will be reduced (e.g. less
contribution to cross ventilation within the house). While some doors had larger
gaps this will not make up for smaller gaps – the increase in air flow as the gap
becomes larger follows a law of diminishing returns.
28
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Table 2 shows that there is some variation in door undercut depending on the
room. Within this relatively small sample, it appears that the bedrooms have the
largest undercuts. It is unclear why this may be the case and how representative
this is of the building stock.
Only one dwelling had at least a 10 mm gap for all rooms measured. All others
had at least one room with a gap of less than 10 mm.
Figure 6 Size of door undercuts
Table 2 Distribution of door undercuts for different rooms
Lounge Kitchen
Master
bedroom
Other
bedrooms
Bathroom En-suite WC
Number 19 17 20 33 18 6 13
Mean (mm) 6.2 7.7 12.2 11.3 7.4 8.0 8.6
Min (mm) 1 1 4 2 0 5 2
Max (mm) 10 21 20 22 20 13 15
St Dev (mm) 3.4 5.6 4.6 4.6 5.5 2.8 4.5
6.4 Whole house air exchange rates
6.4.1 Introduction
This section provides a summary of the results of the air exchange rates. For
results of individual dwellings see Table A.4 in the Appendix.
Chapter 6 Results
|
29
6.4.2 PFT measurements
Figure 7 shows the distribution of air exchange rates measured in the dwellings
using the PFT technique. Figure 8 shows the air exchange rates for both flats and
other dwellings. The mean air exchange rate for the flats was 0.28 ach whereas it
was 0.51 ach for houses.
Figure 9 compares the ventilation rate monitored for each dwelling against the
background ventilation rates recommended in ADF Table 1.1b. This analysis takes
into account that the recommended background ventilation rate depends on
the size of the dwelling and its intended occupancy. None of the flats achieved
the ADF recommended background ventilation rates while 60% of the other
dwellings did.
A number of factors have been identified to help explain these low results:
i. The trickle ventilator areas are sized in ADF 2006 for the winter period. ADF
highlights that due to lower temperature differences between inside and
outside during the spring and autumn periods, thereby lowering the natural
ventilation driving forces through the trickle ventilators, window opening
may be required.
ii. Overall, as shown in Section 6.3, there is under-installation of trickle ventilator
area in these dwellings. On average, only 94% of the required area is
installed, based on the equivalent area marked on the trickle ventilators.
Further checking is also required as to how well this marked equivalent area
(based on laboratory tests) represents the actual equivalent area in practice.
iii. Half of the door undercuts are below the recommended size. This will have
particular significance if the occupant closes their internal doors.
iv. Many of the flats were classed as multi-sided ventilation but, in fact, most of
the ventilation was on one side of the dwelling and limited ventilation along
a second side. Hence, effectively they were single-sided properties and ADF
2006 recommends additional ventilation for such properties.
v. The PFT approximates the air exchange rate. If the ventilation rate varies
much during the measurement period, the approach will tend to under-
read the true air exchange rate. This variation depends on the natural
ventilation driving forces and occupant ventilation behavior. Based on some
computational modeling work undertaken for this study, it is estimated that
the air exchange rates may have under-read by between 5–15%, with the
former for two-storey dwellings and the latter for flats. The results suggest
that there should be greater variation in ventilation rates for flats as they are
reliant on just wind for ventilation whereas for two-storey dwellings they are
reliant on both wind and stack forces which smoothes out variations in the
ventilation rate.
30
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
This is further discussed in the next section when the results are all considered
together.
Figure 7 Air exchange rate estimated using the PFT technique
Figure 8 Air exchange rate for individual dwellings
Chapter 6 Results
|
31
Figure 9 Air exchange rate estimated using the PFT technique vs ADF 2006
6.5 Intermittent extract flow rates
6.5.1 Introduction
Mechanical extract flow rates were measured wherever practically possible. The
following is a summary of the findings. A complete set of results is provided in
Table A.5 of the Appendix.
6.5.2 Kitchen mechanical extract
Figure 10 shows the distribution of the measured flow rates for the 13 dwellings
monitored with an extract over the cooker (all exhausted to the outside). The
mean extract flow rate for these fans was 23 l/s, with two achieving the ADF
recommended flow rate of 30 l/s. Allowing for a 5% error in the measurement
technique (ie pass rate at 28 l/s), six of the 13 dwellings passed.
The minimum flow rate recorded was 9 l/s. In this case, the fan had been replaced
with a quieter fan at the occupants’ request. The quieter model is likely to have
had a lower fan speed and therefore flow rate.
32
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Figure 10 Kitchen extract above the cooker
Figure 11 shows the distribution of the measured flow rates for the five dwellings
monitored with an extract that was not directly above the cooker (all exhausted to
the outside). ADF recommends that these fans have a higher flow rate of 60 l/s as
they are less effective in removing cooking pollutants (fans above or close to the
cooker can remove some cooking pollutants before they become mixed with the
room air). All five fans were measured with a flow rate between 33–35 l/s, thus
significantly below the value recommended in ADF. It is possible that they were
installed to meet the previous 30 l/s criterion.
Figure 11 Kitchen extract not above the cooker
6.5.3 Bathroom mechanical extract
Figure 12 shows the distribution of the measured extract rates in the main
bathrooms of 21 dwellings (not practical to measure the remaining dwelling)
Chapter 6 Results
|
33
and en-suite bathrooms of eight dwellings. ADF recommends that extract rates
should achieve 15 l/s. Only eight (just over one-quarter) of the sample met this
criterion. Allowing for a 5% experimental error (pass rate of 14 l/s), allowed a
further two fans (10 in total) to pass. There appeared to be no difference between
the performance of main and en-suite bathroom fans.
Figure 12 Bathroom extract rates
6.5.4 WC mechanical extract
Figure 13 shows the distribution of the measured flow rates for all 12 dwellings
that had a small WC with an extract fan. ADF recommends that extract rate
should achieve 6 l/s. Seven of the fans achieved this criterion.
34
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Figure 13 Small WC extract rates
6.5.5 Discussion
A significant number of mechanical extract fans were not providing the
recommended flow rates. One possible reason is that the incorrect fan rating
may have been installed. Another potential cause is inadequate installation. The
study was not focused on investigating installation issues. On several occasions,
where the householder had sufficient time and there was easy access, the ducting
installation was viewed. Problems noted included unnecessarily long and flexible
duct runs and a high number of bends.
6.6 Relative humidity
This section provides a summary of the relative humidity (RH) results measured
in living room, kitchen, bathroom and master bedroom for the main sample
(RH measurements were not taken in the kitchen and bathroom of pilot study
homes). For results of individual dwellings see Table A.6 and Table A.7 in the
Appendix.
The consultation version of ADF 2010 proposes new RH criteria to be met during
the heating season in each room:
º average monlhly PH < 65°
º average weekly PH < 75°
º average daily PH < 85°
Chapter 6 Results
|
35
Figure 14 shows the distribution of the maximum average weekly RH level
monitored in each dwelling. Within each dwelling, the results for each room
were determined and the maximum chosen. The maximum average weekly level
recorded was 71% which is within the new criterion of 75%.
If it is assumed that the average weekly level is representative of the average
monthly level, then four homes recorded RH levels of 65% or greater (H11,
H15, H16, H20). Three of these dwellings are flats (H11, H15 and H20), and as
previously noted in Section 6.4, none of the flats achieved the ADF recommended
background ventilation rate. H16 had the lowest air exchange rate recorded in
this study (0.2 ach). It seems reasonable that for these four homes the weekly
levels would be similar to the monthly level, as the external levels were close to
the average in the sample and the occupants were asked to undertake moisture
generating activities (e.g. cooking and bathing) as usual.
Figure 15 shows the distribution of the maximum average daily RH level
monitored in each dwelling. The maximum average daily level recorded was 82%
which is within the new criterion of 85%.
Some additional analysis was undertaken to indicate the likely impact to the
internal RH levels from reasonably worst external RH levels. This is because
internal levels are very dependent on external levels.
The following steps were undertaken:
i. The average weekly vapour pressure level was determined for each room
in each dwelling over the week monitoring period from the measured
temperature and RH levels. This was similarly undertaken for the external
vapour pressure level.
ii. The vapour pressure excess was determined for each room by subtracting the
vapour pressure outside from the vapour pressure in the room.
iii. The reasonably worst case vapour pressure level for outside was determined.
The London TRY (Test Reference Year) data set was used for external climate
conditions. We determined the maximum weekly vapour pressure level over
the heating season (SAP 2005 runs suggested that daily average external
temperature should be below 10.5 °C during the heating season).
iv. The worst case weekly vapour pressure level from (iii) was added to the excess
vapour pressure level from (ii).
v. This was reconverted back to RH level for each room.
The results showed four homes with weekly average levels above 70%, but only
one (H8) exceeded 75% (achieved 75.7%). If necessary, windows can be opened
36
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
to remove excess moisture due to periods of high external levels. Further analysis
could usefully investigate the impact of external moisture levels on the daily
average and whether the TRY data-set is the most appropriate one to use for such
analysis.
Figure 14 Highest weekly average RH level in each dwelling
Figure 15 Highest daily average RH level in each dwelling
Chapter 6 Results
|
37
6.7 Nitrogen dioxide
This section provides a summary of the results of the nitrogen dioxide (NO
2
) levels.
For results of individual dwellings see Table A.8 in the Appendix.
Figure 16 provides the distribution of the average NO
2
levels measured in the
kitchen over the monitoring period. The mean concentration of NO
2
recorded in
the sample was 24 μg.m
-3
. This increased to 29 μg.m
-3
if you consider only those
64% of dwellings that used gas for cooking.
Four dwellings exceeded the recommendation in ADF that NO
2
levels should not
exceed 40 μg.m
-3
as a long term average.
º Three homes were inslalled wilh lans lhal were nol adjacenl lo lhe cooker
and achieved approximately 50% of the recommended flow rate. Hence, this
is a plausible explanation for the high NO
2
levels.
º Wilhin lhe lourlh home, lhe occupanls did nol use exlracl venlilalion during
cooking even though requested to do so. The occupant commented in the
daily diary that the noise level of the fan was a nuisance, therefore the fan
was not used.
Figure 16 Distribution of NO
2
concentrations
38
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
6.8 TVOCs
This section provides a summary of the results of the TVOC levels measured in the
living room and master bedroom. For results of individual dwellings see Table A.9
in the Appendix. There was an error in the analysis of a batch of TVOC sample
tubes which voided results for seven of the dwellings.
Figure 17 provides the distribution of the highest TVOC levels monitored in each
home. 53% were higher than the guideline figure of 300 μg.m
-3
in ADF 2006.
Three of these dwellings had levels between 300–330 μg.m
-3
. Given the
accuracy of the detection technique, these levels could actually have been below
300 μg.m
-3
(or even higher). However, even if just below 300 μg.m
-3
, they are
approaching this guideline value and hence still a cause for concern.
Reviewing the data, overall there is no clear indication of a particular source or
sources of TVOCs that is causing this high level. For one of the dwellings above
600 μg.m
-3
(H7), the occupant did smoke in the main bedroom (the room of the
highest reading) which may have significantly impacted on the results directly or
indirectly through the use of any odour-masking chemical products. However,
smoking also took place in three other dwellings (H5, H6, H8) and the results did
not show this significant rise. Hence it is not clear whether smoking is the cause.
For the other dwelling above 600 μg.m
-3
, there was no obvious source from the
completed questionnaire and the occupants had already moved by the time we
followed up after the monitoring period.
Figure 17 Highest TVOC concentration in each dwelling
Chapter 6 Results
|
39
6.9 Formaldehyde
This section provides a summary of the results of the formaldehyde (HCHO)
levels measured in the living room and master bedroom. For results of individual
dwellings see Table A.10 in the Appendix.
Figure 18 provides the distribution of the highest formaldehyde levels monitored
in each home. All results were less than the WHO guideline level for effects on
health and comfort of 100 μg.m
-3
(WHO 2000) averaged over 30 minutes.
Formaldehyde levels monitored during the study were averaged over three
days. The highest level recorded during the monitoring period was 63 μg.m
-3
in a master bedroom. It would be expected that there would be some variation
in these three days such that there would be 30-minute periods that would be
higher. Assuming the sources are predominantly formaldehyde based resins,
e.g. in wood based products, then the rate of emission will be quite steady,
although somewhat elevated at higher temperatures and humidities. Pollutant
concentrations are more likely to vary due to variations in ventilation rate – there
are likely to be some periods of significantly lower ventilation rate due to falls in
external driving forces. Assuming that this would result in maximum 30-minute
levels double that of the three-day readings, it would suggest that at most three
dwellings would have exceeded the WHO criterion. It is interesting to note that
the three highest levels corresponded to the three dwellings with the lowest air
permeability levels (H7, H18, H20).
Figure 18 Highest formaldehyde concentration in each dwelling
40
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Chapter 7
Discussion
7.1 Are intermittent extract flow rates sufficient?
The purpose of intermittent extract ventilation is to provide extract ventilation
in rooms where there are significant indoor pollutant sources. This is to both
minimise the pollutant levels in those rooms and to minimise the spread of
pollutants to the rest of the building. Extract ventilation is located in the kitchen
to remove combustion pollutants from cooking and other moisture generating
activities, e.g. washing-up. Extract ventilation is also located in bathrooms to
remove moisture generated in the air from bathing and showering.
The results from this study suggest that the intermittent extract flow rates may
be insufficient to control moisture levels generated in the wet rooms. Excessive
levels were recorded in the wet rooms of three homes (H15, H16, H20) against
the monthly RH criterion. In each case, the exceedance occurred in the bathroom
and not the kitchen. This analysis is based on the reasonable assumption that the
actual monthly levels in these homes were similar to the recorded weekly levels.
However, the results may be at least partially explained by the low air exchange
rates in these homes. In two of the homes, the bathroom extract rates actually
met CLG recommended levels (H15 and H20). The air exchange rate in all three of
these homes was low (0.20 – 0.24 ach) and would tend to generally increase the
relative humidity level throughout the home.
Based on the same assumptions, the monthly RH levels were also high in the
habitable rooms of three homes. These occurred in the living room for H16 and in
the bedroom for H11 and H20.
Again, low exchange rates are expected to have a significant influence on these
results. RH levels in habitable rooms are controlled by a combination of extract
ventilation to minimise dispersal from wet rooms to the rest of the dwelling and
background ventilation to remove moisture produced elsewhere in the dwellings,
for example through respiration. Insufficient extract ventilation is likely to be one
factor allowing dispersal to the rest of the home. However, in all three homes it is
noted that the air exchange rate is low (0.20 – 0.27 ach).
Chapter 7 Discussion
|
41
Further analysis considered a reasonably worst case external moisture scenario.
One dwelling (H8) exceeded the 75% RH ADF 2010 criterion for weekly average
indoor RH levels and three other dwellings approached this level (H2, H15, H20).
In these cases, either the intermittent extract could have been used for a longer
period or windows could have been opened for short periods. It is worth noting
that of the four dwellings, three had air permeability below 5 m
3
.h
-1
.m
-2
, although
the sample size is very small to make any conclusive judgement based on this. It is
also worth noting that these highest levels always occurred in the bathroom.
It is perhaps surprising that the RH levels were not higher more generally, as
in many cases intermittent extract ventilation and/or trickle ventilator areas
did not meet ADF 2006 guidance. Focusing particularly on the wet rooms
here (background ventilation is discussed further on), there are two plausible
explanations.
º The exlracl low rales were chosen lor ADl based on reasonably worsl
cases moisture generation rates. These may not have occurred in practice
because of user behaviour (e.g. lower moisture production during cooking
than predicted) or relatively low occupancy. (ADF assumes two people sleep
in the master bedroom and one person sleeps in each other bedroom, but,
in practice, all apart from one of the dwellings had at least one bedroom
unoccupied during the monitoring period.)
º The occupanls may nol have closed lhe inlernal doors lo lhe wel rooms
during moisture production. Hence, the moisture levels would have been
reduced in these rooms but spread and diluted throughout the rest of the
dwelling.
The intermittent extract flow rates were not sufficient to control nitrogen
dioxide levels. In four dwellings, levels exceeded those recommended by ADF.
However, as discussed in Section 6.7, three of the dwellings had extract flow rates
approximately half that recommended by ADF and if the fans had been properly
installed this should have addressed this problem. A further dwelling did not
use their cooker hood as it was too noisy. Whilst it is important to recognise this
problem, it does suggest that if the cooker hood had been used, nitrogen dioxide
levels would have been controlled.
Overall the results from this study suggest that the intermittent extract flow
rates are sufficient and should not be changed. This includes the more airtight
dwellings in the sample. However, it is important that cooker hoods and extract
fans are properly installed.
42
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
7.2 Are trickle ventilator areas sufficient?
7.2.1 Introduction
The purpose of background ventilation, of which trickle ventilators are a means to
achieving it, is to remove non-localised pollutants spread around the dwelling as
well as remove pollutants that ‘escape’ extract ventilation which, in practice, can
never be 100% effective (e.g. if the kitchen door is open during cooking, some
cooking pollutants will be extracted by the fan and others spread to the rest of the
dwelling through the open door).
Of all of the pollutants monitored in the habitable rooms, the RH and TVOC levels
exceeded consultation ADF 2010 pollutant criteria. The formaldehyde levels from
sources around the home were sufficiently low. There may be problems from
formaldehyde in airtight homes – this is a relatively small study and care should be
taken in extrapolating the results to the whole new building stock. However, by
adequately controlling RH and TVOCs, it may significantly control other pollutants
as well.
We shall first review the correlation of pollutants with airtightness. It is worthwhile
reviewing this relationship given the purpose of the study.
We shall next review whether TVOC levels are high because of inadequate
installation and inspection or because the ventilation areas recommended in ADF
2006 are insufficient. In doing so, some approximations will need to be made as
it is difficult to know exactly the impact of inadequate installation and inspection.
This analysis focuses on TVOC levels, rather than RH levels, as there was a greater
prevalence of TVOC levels exceeding the IAQ criteria. It is assumed that if TVOC
levels were adequately controlled, this should suffice for RH levels as well.
7.2.2 Pollutant levels versus airtightness
We first review the evidence for a correlation between the air permeability and
indoor pollutant levels. Due to the relatively small sample size, we have not
undertaken detailed statistical analysis.
Table 3 compares air permeability against pollutant levels. The table has been
ordered from lowest to highest air permeability. The table focuses on pollutant
levels in habitable rooms (ie extract fans should address high pollutant levels
produced intermittently in wet rooms).
The pollutants’ levels are colour coded:
º lhe highesl lhird in red
º lhe middle lhird in yellow
º lhe lowesl lhird in green.
Chapter 7 Discussion
|
43
The table suggests, given the limitations of the sample size, the anticipated
correlation between air permeability and pollutant levels. Relatively high levels are
observed in the five most airtight dwellings.
However, there may be a confounding factor here. Four of the five more airtight
dwellings are flats. As highlighted in Section 6.4, it is thought that the flats may
be significantly under-ventilated due to them having effectively a single façade
but the ventilation system being designed for a multi-sided façade.
Table 4 still suggests that even with the flats removed, there is a correlation
between air permeability and indoor pollutant levels. However, it is less clear.
It is not surprising that there is not a clear correlation between airtightness and
indoor pollutant levels. For example, pollutant emission rates will vary between
dwellings. Furthermore, as highlighted earlier, the installed capacity of the
ventilation systems shows significant variation.
Table 3 Comparison of air permeability and pollutant level
Property details Air permeability Ventilation RH (habitable rooms) TVOCs HCHO
ID House type
Test
Method
A
Test
Method
C
Weekly
average
Worse
weekly
average
Max Max
m
3
.h
-1
.m
-2
@ 50 Pa
m
3
.h
-1
.m
-2
@ 50 Pa
Ach % RH % RH μg.m
-3
μg.m
-3
H20 Flat 2.96 4.32 0.24 71.0 74.6 535 50
H15 Flat 3.37 4.92 0.23 61.2 73.6 n/a 55
H11 Flat 3.45 4.32 0.27 65.0 69.3 n/a 37
H7 Terrace-End 3.89 6.25 0.44 53.8 66.5 976 63
H18 Flat 4.1 5.02 0.42 58.7 63.7 1213 35
H1 Detached 4.12 6.22 0.55 50.4 62.4 224 30.2
H19 Flat 4.52 5.09 0.23 49.9 61.2 572 28.1
H8 Terrace-Mid 4.89 7.22 0.62 58.1 75.7 311 22.3
H12 Semi-Det 4.96 7.35 0.99 56.0 66.5 n/a 19.6
H5 Terrace-End 4.98 7.81 0.26 53.9 67.5 111 16.4
H6 Terrace-Mid 5.13 7.37 0.70 46.8 n/a 204 29
H10 Terrace-Mid 5.37 7.82 0.38 42.5 58.2 191 11.7
H9 Terrace-End 5.49 8.25 0.37 46.0 65.9 330 31
H2 Detached 6.45 9.77 0.47 62.2 73.5 464 31
H14 Semi-Det 7.02 9.67 0.43 54.5 63.8 n/a 32
H16 Semi-Det 7.18 8.67 0.2 66.3 n/a n/a 46
H3 Detached 8.28 11.53 0.49 43.1 57.8 156 19.9
H13 Detached 8.58 13.11 0.97 41.8 56.6 n/a 13.1
H17 Detached 8.84 11.3 0.45 52.3 56.7 n/a 24.8
P1 Semi-Det 9.51 13.3 0.38 n/a n/a 53 15.5
P2 Semi-Det 10.61 13.74 0.34 n/a n/a 231 19
H4 Detached 11.6 14.9 0.71 39.8 55.9 320 24.5
44
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Table 4 Comparison of air permeability and pollutant level (without flats)
Property details Air permeability Ventilation RH (habitable rooms) TVOCs HCHO
ID House type
Test
Method
A
Test
Method
C
Weekly
average
Worse
weekly
average
Max Max
m
3
.h
-1
.m
-2
@ 50 Pa
m
3
.h
-1
.m
-2
@ 50 Pa
Ach % RH % RH μg.m
-3
μg.m
-3
H7 Terrace-End 3.89 6.25 0.44 53.8 66.5 976 63
H1 Detached 4.12 6.22 0.55 50.4 62.4 224 30.2
H8 Terrace-Mid 4.89 7.22 0.62 58.1 75.7 311 22.3
H12 Semi-Det 4.96 7.35 0.99 56.0 66.5 n/a 19.6
H5 Terrace-End 4.98 7.81 0.26 53.9 67.5 111 16.4
H6 Terrace-Mid 5.13 7.37 0.70 46.8 n/a 204 29
H10 Terrace-Mid 5.37 7.82 0.38 42.5 58.2 191 11.7
H9 Terrace-End 5.49 8.25 0.37 46.0 65.9 330 31
H2 Detached 6.45 9.77 0.47 62.2 73.5 464 31
H14 Semi-Det 7.02 9.67 0.43 54.5 63.8 n/a 32
H16 Semi-Det 7.18 8.67 0.2 66.3 n/a n/a 46
H3 Detached 8.28 11.53 0.49 43.1 57.8 156 19.9
H13 Detached 8.58 13.11 0.97 41.8 56.6 n/a 13.1
H17 Detached 8.84 11.3 0.45 52.3 56.7 n/a 24.8
P1 Semi-Det 9.51 13.3 0.38 n/a n/a 53 15.5
P2 Semi-Det 10.61 13.74 0.34 n/a n/a 231 19
H4 Detached 11.6 14.9 0.71 39.8 55.9 320 24.5
7.2.3 Review of TVOC levels
First, it is informative to review the TVOC levels in more detail. The levels are
shown in Table 5, compared to other parameters. The data have been sorted by
TVOC level.
The Table highlights the following:
º lhose air permeabilily levels wilh Melhod A below 5 m
3
.h
-1
.m
-2
@ 50Pa
º lhe lve lowesl levels ol Melhod C and venlilalion.
As before, from this limited sample size, there is a suggestion that the high TVOC
levels occur in more airtight properties. The five of the eight properties with high
TVOC levels had air permeabilities below 5 m
3
.h
-1
.m
-2
. Perhaps of more concern
is that the other three were leakier and, if they had been more airtight, the
ventilation would be expected to be less and consequently the TVOC levels would
have been higher.
Chapter 7 Discussion
|
45
As referred in the previous sub-section, not all airtight properties are expected
to have high pollutant levels, e.g. it depends on the presence of sources. This is
a possible explanation for the three dwellings, with an air permeability below
5 m
3
.h
-1
.m
-2
, having acceptable VOC levels.
It is also noted that three of the four highest level homes are flats which is
explored further below.
Finally, it is important to note that TVOCs were chosen in ADF to represent organic
compounds as it was to be the most sensitive marker for organic compounds.
If TVOCs are less than the guideline level of 300 μg.m
-3
, then typically individual
organic compounds (e.g. formaldehyde, toluene, benzene) should also be below
health-based guidelines for these specific compounds.
Table 5 Review of TVOC levels
House
ID
House type TVOC
Percentage
above
300 μg.m
-3
Method A Method C Ventilation
μg.m
-3
%
m
3
.h
-1
.m
-2
@
50 Pa
m
3
.h
-1
.m
-2
@
50 Pa
Ach
P1 Semi-Det 53 0 9.5 13.3 0.38
H5 Terrace-End 111 0 5.0 7.8 0.26
H3 Detached 156 0 8.3 11.5 0.49
H10 Terrace-Mid 191 0 5.4 7.8 0.38
H6 Terrace-Mid 204 0 5.1 7.4 0.7
H1 Detached 224 0 4.1 6.2 0.55
P2 Semi-Det 231 0 10.6 13.7 0.34
H8 Terrace-Mid 311 4 4.9 7.2 0.62
H4 Detached 320 7 11.6 14.9 0.71
H9 Terrace-End 330 10 5.5 8.3 0.37
H2 Detached 464 55 6.5 9.8 0.47
H20 Flat 535 78 3.0 4.3 0.24
H19 Flat 572 91 4.5 5.1 0.23
H7 Terrace-End 976 225 3.9 6.3 0.44
H18 Flat 1213 304 4.1 5.0 0.42
7.2.4 Impact of installing inadequate trickle ventilation
This next step is to estimate the impact on the TVOC levels from the fact that
insufficient trickle ventilation was installed in some homes. If sufficient trickle
ventilation was installed, how would the TVOC levels vary?
46
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
In doing this analysis, we make a number of assumptions.
º lirslly, we assume lhal lhe venlilalion rale is proporlional lo lhe air leakage
rate at 50 Pa. This assumption is the basis of the 1/20th rule of thumb (see
Sherman, 1987
2
).
Hence:
Ventilation = k. ACH
50
where k = the proportionality constant.
We already know the air pressure test results with and without the trickle
ventilation open. We can then estimate the increase in ventilation rate, if the
correct amount of trickle ventilation was included as follows.
Ventilation
corrected
= k. ((
trickle vent (ADF)
x [ACH
C
– ACH
B
]) + ACH
B
)
trickle vent (installed)
where ACH
B
and ACH
C
are the air leakage rate results at 50 Pa for Methods B and
C, respectively.
º Second, we assume lhal lhe venlilalion rale is inversely proporlional lo
the TVOC pollutant concentration (ie ventilation rate = 1/TVOC level).
This is reasonable. It is the standard formula for pollutant concentration in
equilibrium. It assumes that the outdoor concentration has a small impact on
the internal levels which is suggested by the results in this and other studies.
Taking these two assumptions, the results are shown in Table 6. In this analysis,
the TVOC levels are reduced where there was previously under-installed trickle
ventilation. For comparison, TVOC levels are increased where previously there
was an excessive trickle ventilation rate compared to ADF 2006. Note that the
analysis was not possible for dwellings P1, P2, H9 and H10 due to a lack of
equivalent area information on the trickle ventilator to determine the degree of
any under or over installation.
Overall, while this does have an impact, it is not significantly reducing those
homes with the highest TVOC levels. H4 and H8 are now less than or equal to
300 μg.m
-3
, but this is still a cause for concern as they are still approaching the
guideline level (and, could be expected on a different week’s monitoring period,
to exceed it). Even if the above assumptions are underestimates, doubling any
reduction noted here would still not bring the high TVOC results significantly
below 300 μg.m
-3
, if at all.
2
Sherman, M H (1987) Estimate of infiltration from leakage and climate indicators. Energy and Buildings, 10, pp. 81–86.
Chapter 7 Discussion
|
47
Table 6 Analysis of TVOC results due to incorrectly installed trickle ventilation
capacity
House ID House type
Method
A
TVOC
(original)
TVOC
(new)
Change
m
3
.h
-1
.m
-2
@ 50 Pa
μg.m
-3
μg.m
-3
%
P1 Semi-Det 9.5 53
P2 Semi-Det 10.6 231
H1 Detached 4.1 224 185 -17
H2 Detached 6.5 464 488 +5
H3 Detached 8.3 156 172 +10
H4 Detached 11.6 320 300 -6
H5 Terrace-End 5.0 111 109 -2
H6 Terrace-Mid 5.1 204 198 -3
H7 Terrace-End 3.9 976 890 -9
H8 Terrace-Mid 4.9 311 296 -5
H9 Terrace-End 5.5 330
H10 Terrace-Mid 5.4 191
H18 Flat 4.1 1213 1126 -7
H19 Flat 4.5 572 530 -7
H20 Flat 3.0 535 472 -12
7.2.5 Building more airtight dwellings
It is also interesting to see the potential impact if all of the dwellings were built
at least to 4 m
3
.h
-1
.m
-2
. Hence for those properties leakier than this, the impact
of a more airtight property was determined applying the same techniques as in
Section 7.2.4. This is shown in Table 7 below.
The results suggest very significant increases in TVOC levels. If all dwellings had
an air permeability built to 4 m
3
.h
-1
.m
-2
or better, nine homes would now exceed
300 μg.m
-3
.
It should also be noted that this would also impact on the other pollutant levels.
º Similar percenlage increases could be expecled lor lormaldehyde. However,
analysis showed that this does not tend to take the levels above 100 μg.m
-3
.
The highest formaldehyde levels were for the more airtight dwellings
and hence only a small increase, if at all, is observed by assuming an air
48
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
permeability of 4 m
3
.h
-1
.m
-2
. For those leakier dwellings, the levels were fairly
low to start with and, even with increases, the results were still relatively low.
º An increase would also be expecled in moislure levels. However, lhe
percentage increase would be less than that for TVOCs and formaldehyde.
Indoor moisture levels arise from both indoor and external sources. To a first
approximation, the external sources provide a background level whatever the
air permeability. The air permeability has a greater impact on internal sources
– an improved air permeability stops moisture from escaping the dwelling
and, therefore, increasing the indoor levels.
Table 7 Impact of improving airtightness on TVOC levels
House ID House type Method A
TVOC
(original)
TVOC
(new)
Change
m
3
.h
-1
.m
-2
@ 50 Pa
μg.m
-3
μg.m
-3
%
P1 Semi-Det 9.5 53 91 +71
P2 Semi-Det 10.6 231 445 +93
H1 Detached 4.1 224 228 +2
H2 Detached 6.5 464 620 +33
H3 Detached 8.3 156 248 +59
H4 Detached 11.6 320 653 +104
H5 Terrace-End 5.0 111 127 +14
H6 Terrace-Mid 5.1 204 241 +18
H7
*
Terrace-End 3.9 976
H8 Terrace-Mid 4.9 311 355 +14
H9 Terrace-End 5.5 330 403 +22
H10 Terrace-Mid 5.4 191 231 +21
H18 Flat 4.1 1213 1237 +2
H19 Flat 4.5 572 637 +11
H20
1
Flat 3.0 535
*
These dwellings already had an air permeability better than 4 m
3
.h
-1
.m
-2
, so no analysis was undertaken
7.2.6 Inadequate ADF guidance in flats
There is a suggestion that a contributing factor in flats is inadequate guidance
in ADF. Flats 18 to 20 had two external façades, hence the trickle ventilation
design reflected this. However, they were essentially single sided, in that one
of the façades was short in length and thus little cross ventilation would occur
Chapter 7 Discussion
|
49
in practice. It may have been better if the ventilation guidance for single-sided
façades was followed or another ventilation system adopted.
It is difficult to assess the impact here. Assuming that better guidance would
result in a doubling of the ventilation rate from purpose-provided ventilation,
according to similar calculation as in Section 7.2.4, this would reduce the TVOC
levels by a further 15–30%. The results would be for H18–20: 893, 444 and
331 μg.m
-3
, respectively.
7.2.7 Impact of door undercuts
This is to consider the impact of undercuts being less than recommended by
ADF2006. Just over 50% of the door undercuts did not achieve 10 mm and only
one dwelling had all of its door undercuts achieving this target.
As shown in Table 8, there appears no clear correlation between door undercuts
and TVOC levels. Door undercuts were not measured in the pilot homes and
hence these results are not presented here.
This lack of a clear correlation can be explained by several factors. Again, there
will be different source emission rates in different homes. Furthermore, the
impact of door undercuts depends on whether internal doors are left open
and whether there are any other routes for air to move around the dwelling
(i.e. internal leakage through gaps, cracks separating internal rooms).
50
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Table 8 Impact of door undercuts
House ID House type TVOC
Undercuts
(all rooms)
Undercuts
(habitable
rooms only)
μg.m
-3
mm mm
H5 Terrace-End 111 9 6
H3 Detached 156 16 16
H10 Terrace-Mid 191 9 14
H6 Terrace-Mid 204 4 5
H1 Detached 224 11 13
H8 Terrace-Mid 311 7 7
H4 Detached 320 12 13
H9 Terrace-End 330 7 7
H2 Detached 464 11 13
H20 Flat 535 7 10
H19 Flat 572 9 12
H7 Terrace-End 976 5 5
H18 Flat 1213 9 11
7.2.8 Discussion
From this study, it is not possible to provide conclusive evidence that dwellings
built to 3-4 m
3
.h
-1
.m
-2
or tighter (ie those homes for which ADF 2006 does not
provide guidance) need additional ventilation compared to that recommended
in ADF 2006. This is mainly due to the need to estimate for effects of inadequate
installation and inspection. Furthermore, in the case of flats, better guidance is
needed for single-sided ventilation.
However, overall, there is the suggestion that additional ventilation is needed for
more airtight dwellings for the following reasons.
º Lighl ol lhe !5 homes wilh TVOC measuremenls had levels above lhe level ol
300 μg.m
-3
recommended in ADF 2006.
º The projecled impacl on TVOC levels ol lhe dwellings in lhis sample being
tightened to 4 m
3
.h
-1
.m
-2
appears to significantly outweigh likely effects from
under installation of trickle ventilation. Furthermore, it is likely to result in nine
of the 15 homes with TVOC levels above 300 μg.m
-3
.
º Three ol lhe dwellings wilh lhe highesl TVOC levels are lals. These may
have been significantly under-ventilated as they were designed for two-
Chapter 7 Discussion
|
51
sided ventilation but ventilation was effectively only single-sided. It is not
clear whether better design would have reduced levels below 300 μg.m
-3
.
However, this still leaves five to six dwellings with levels above 300 μg.m
-3
.
º ll is anlicipaled lhal having all door underculs al leasl !0 mm would help.
However, it is unclear from the results the benefit that this would have.
There is an additional confounding factor. It is possible that the trickle ventilators,
while producing the equivalent area in the laboratory, may not be achieving this
on installation taking into account their actual surroundings which could impede
flow. This is something that should be considered in further work.
Finally, it is important to note that in practice, occupants do not use their
ventilation system to full capacity (trickle vents are not always open and fans are
not always used). Hence, the actual pollutant levels in dwellings will likely be
higher than those recorded in this study.
7.3 Are dwellings Part F compliant?
All dwellings tested are Part F compliant in that they have been given approval
through the building control process.
However, the study has demonstrated many cases where the recommendations
given in ADF 2006 have not been met. The key differences from the ADF
recommendations are as follows:
º Many inlermillenl exlracl low rales do nol meel lhose recommended in
ADF. Every dwelling had at least one fan or cooker hood that did not provide
the recommended extract rate. It is thought that this resulted in, at least, the
nitrogen dioxide levels exceeding WHO guidelines in at least three dwellings.
º 72° ol dwellings had lolal lrickle venlilalor area below lhal recommended
in ADF.
º Only one dwelling had all door underculs ol !0 mm or grealer.
7.4 Are dwellings Part L compliant?
All dwellings tested are Part L compliant in that they have been given approval
through the building control process.
Two homes exceeded the maximum level of 10 m
3
.h
-1
.m
-2
required by ADL 2006
for dwellings sampled. It is not possible to know if these dwellings exceeded
10 m
3
.h
-1
.m
-2
on completion, as they may not have been included in the
52
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
developer’s air permeability test sample. It is feasible that the dwellings achieved
10 m
3
.h
-1
.m
-2
on completion and the air permeability increased post-completion
due to drying out of the properties and/or work undertaken by the occupants.
7.5 Should we change air pressure testing for Part L
compliance?
These results question the need to move from BS EN 13829 Method A to Method
B. There is relatively little difference between the two sets of results (mean of
0.2 m
3
.h
-1
.m
-2
; maximum of 0.5 m
3
.h
-1
.m
-2
).
The advantages of making this change are as follows:
º ll provides a more accurale assessmenl ol lhe leakage ol lhe conslruclion
of the dwelling. This will become more important as air permeability further
improves and the relative contribution from non-sealed trickle ventilation
increases.
º ll allows lor lhe lacl lhal some lrickle venlilalors are designed nol lo close
to ensure that there is some ventilation at their minimum setting. This may
become more prevalent as homes become more airtight due to concern
that occupants will close trickle ventilation in airtight homes because of
inadequate information and result in poor indoor air quality and potential
health effects.
This needs to be balanced by the additional time and effort to seal and unseal all
natural ventilation openings given the relatively small air permeability reduction.
7.6 Should we undertake air pressure testing for Part F
compliance?
The study investigated whether it was possible to determine the total trickle
ventilation equivalent area through testing in-situ. The proposal was to undertake
air pressure testing with the trickle ventilators sealed and then opened and
through analysis of the resultant data determine the equivalent area.
This study has not demonstrated the robustness of the approach. The approach,
at present, is not able to determine the equivalent area with sufficient accuracy.
However, further development of this approach (e.g. undertake air pressure tests
at lower pressure differences) may make it a useful technique.
Chapter 8 Conclusions and recommendations
|
53
Chapter 8
Conclusions and recommendations
The key results from this study can be summarised as follows:
º We have monilored 22 dwellings buill lo Parls l and L 2006, wilh lrickle
ventilation and intermittent extract. Ten of the 22 dwellings tested (46% of
total sample including pilot homes, 50% of main sample) achieved an air
permeability of less than 5 m
3
.h
-1
.m
-2
.
º Nearly lhree-quarlers ol lhe dwellings (72° ol sample) did nol have
sufficient trickle ventilation installed to meet ADF 2006. In the worst case,
only just over half (57%) of the recommended trickle ventilation was
installed.
º 52° ol lhe inlernal doors lailed lo achieve lhe !0 mm gap under lhe doors
recommended in ADF 2006.
º Less lhan hall ol lhe kilchen or balhroom lans achieved lhe low rales
recommended in ADF 2006. Furthermore, none of the fans located away
from the cooker achieved the recommended flow rates in the kitchen.
º All relalive humidily levels monilored over a week-long period were wilhin
the new daily and weekly relative humidity guideline levels proposed
in the consultation version of ADF 2010. However, making reasonable
assumptions, it would be expected that approximately four homes would
have monthly relative humidity levels that would exceed the newly proposed
monthly relative humidity guideline level.
º lour homes had levels ol nilrogen dioxide in lhe kilchen lhal exceeded
that recommended in ADF 2006. A key issue in these cases was insufficient
installed flow where fans are located away from the cooker.
º All homes had acceplable levels ol lormaldehyde.
º Over hall ol lhe homes wilh TVOC measuremenls had levels exceeding
that recommended in ADF 2006. It is important to note that in developing
ADF 2006, TVOCs were chosen to represent organic compounds. A review
of indoor air quality results suggested that if the TVOC guideline level of
300 μg.m
-3
was met, then individual chemical health-based guideline levels
(e.g. benzene, toluene, formaldehyde) should also be met.
54
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
The key conclusions from these results are as follows:
º ll is dillcull lo make a conclusive judgemenl as lo whelher or nol lhe currenl
recommended natural ventilation provisions in ADF 2006 are sufficient
for airtight homes. The study is relatively small and the conclusions should
only be treated as indicative. Furthermore, in all cases, the capacity of the
ventilation system did not meet that recommended in ADF 2006.
º The inlermillenl exlracl rales appear lo be sullcienl. ll is imporlanl lo ensure
that they are installed correctly to provide the capacity recommended in ADF
2006. There is no evidence from this study that with correct installation, the
extract rates should be increased for more airtight dwellings.
º The lrickle venlilalor sizes appear lo be insullcienl. Lven allowing lor
installation issues for the ventilation system as a whole, at least a significant
minority of TVOC levels would be expected to exceed the recommended
guideline level, particularly for the more airtight homes. VOCs are produced
by building products and the activities of the occupants such as smoking,
use of personal hygiene products and interior decorating. It may seem
pedantic to change the design of a ventilation system just to better control
one pollutant, but the long term effects on health due to exposure to
VOCs are not well documented, and it seems prudent to err on the side of
caution. Furthermore, the TVOC criterion was selected as a sensitive marker
for individual organic chemical compounds (ie if TVOC levels are below
the criterion, previous research suggest that individual organic chemical
compounds would also be below recognised indoor or outdoor health-based
levels for these pollutants). There are also other hazards to health that are
affected by ventilation, such as house dust mites, which are associated with
respiratory illnesses including asthma. We do not currently attempt to control
these under Part F because their breeding success is influenced by heating
and hygiene practice as much as ventilation. Improving ventilation would be a
step in the right direction to limit this risk.
º 8eller guidance needs lo be provided lor lhe nalural venlilalion ol lals.
While the flats had two façades, and were ventilated as ‘multi-sided façades’
according to ADF 2006, they were effectively single façades as the second
façade was limited. The ventilation and indoor air quality results suggest that
these homes were under-ventilated.
º ll is imporlanl lo nole lhal in praclice, occupanls do nol use lheir venlilalion
system to full capacity (trickle vents are not always open and fans are not
always used). Hence, the actual pollutant levels in dwellings will likely be
higher than those recorded in this study.
Chapter 8 Conclusions and recommendations
|
55
The key recommendations from this study are as follows:
º ll is imporlanl lhal lhe venlilalion syslem should be inslalled correclly and
inspected to provide the ventilation capacity as designed.
º lurlher evidence needs lo be oblained lo subslanliale lhe indicalors lrom
this study. However, there are sufficient grounds to support the proposal to
increase trickle ventilators in dwellings having an air permeability equal to
or tighter than 4 m
3
.h
-1
.m
-2
. This is based on the levels of TVOCs observed,
and the implications for individual organic compounds. This could also have
benefits in controlling other hazards to health, but this is difficult to quantify.
º This was a relalively small sludy. A larger sludy is necessary, building on lhe
findings of this study, to better determine indoor air quality levels achieved in
airtight dwellings in which the ventilation system is installed correctly.
º Civen lhe relalively high level ol TVOCs, grealer consideralion should be
given to reducing source strength by product controls.
56
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Appendix A
Detailed results
Table A.1 Summary of air permeability test results
Property details Air permeability
ID House type
Test
Method A
Test
Method B
Test
Method C
A-B
Difference
A-B
Difference
m
3
.h
-1
.m
-2
@
50 Pa
m
3
.h
-1
.m
-2
@
50 Pa
m
3
.h
-1
.m
-2
@
50 Pa
m
3
.h
-1
.m
-2
@
50 Pa
%
P1 Semi-Det 9.51 9.14 13.30 0.37 3.9
P2 Semi-Det 10.61 10.18 13.74 0.43 4.1
H1 Detached 4.12 3.98 6.22 0.14 3.4
H2 Detached 6.45 6.23 9.77 0.22 3.4
H3 Detached 8.28 7.81 11.53 0.47 5.7
H4 Detached 11.60 11.19 14.90 0.41 3.5
H5 Terrace-End 4.98 4.88 7.81 0.10 2.0
H6 Terrace-Mid 5.13 4.99 7.37 0.14 2.7
H7 Terrace-End 3.89 3.85 6.25 0.04 1.0
H8 Terrace-Mid 4.89 4.73 7.22 0.16 3.3
H9 Terrace-End 5.49 5.59 8.25 -0.10 -1.8
H10 Terrace-Mid 5.37 5.36 7.82 0.01 0.2
H11 Flat 3.45 3.30 4.32 0.15 4.3
H12 Semi-Det 4.96 4.86 7.35 0.10 2.0
H13 Detached 8.58 8.10 13.11 0.48 5.6
H14 Semi-Det 7.02 6.73 9.67 0.29 4.1
H15 Flat 3.37 3.23 4.92 0.14 4.2
H16 Semi-Det 7.18 6.65 8.67 0.53 7.4
H17 Detached 8.84 8.48 11.30 0.36 4.1
H18 Flat 4.10 3.99 5.02 0.11 2.7
H19 Flat 4.52 4.43 5.09 0.09 2.0
H20 Flat 2.96 2.80 4.32 0.16 5.4
Mean 6.15 5.93 8.54 0.22 3.3
Min 2.96 2.80 4.32
Max 11.60 11.19 14.90
Appendix A Detailed results
|
57
Table A.2 Summary of trickle ventilation equivalent areas
Property details Equivalent area
ID House type Recorded
ADF
recommended
Recorded/
recommended
mm
2
mm
2
P1 Semi-Det N/A
1
35,000
P2 Semi-Det N/A
1
35,000
H1 Detached 53,770 85,000 0.63
H2 Detached 40,500 35,000 1.16
H3 Detached 41,900 30,000 1.40
H4 Detached 75,300 95,000 0.79
H5 Terrace-End 42,500 45,000 0.94
H6 Terrace-Mid 27,500 30,000 0.92
H7 Terrace-End 40,000 50,000 0.80
H8 Terrace-Mid 35,000 40,000 0.88
H9 Terrace-End N/A
2
30,000
H10 Terrace-Mid N/A
2
30,000
H11 Flat 25,470 35,000 0.73
H12 Semi-Det 33,600 45,000 0.75
H13 Detached 64,800 40,000 1.62
H14 Semi-Det 38,500 35,000 1.10
H15 Flat 25,470 35,000 0.73
H16 Semi-Det 41,400 40,000 1.04
H17 Detached 56,700 60,000 0.95
H18 Flat 25,470 35,000 0.73
H19 Flat 21,750 35,000 0.62
H20 Flat 25,470 35,000 0.73
Mean 0.92
Min 0.62
Max 1.62
1
Not measured in pilot study.
2
Total geometric area of 57,300 mm
2
measured in both dwellings.
58
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Table A.3 Summary of trickle ventilation positions (recorded on initial survey)
Property details
Number of installed
trickle vents
Trickle ventilator
position
ID House type Open Closed
P1 Semi-Det 13 0 13
P2 Semi-Det 10 3 7
H1 Detached 21 19 2
H2 Detached 16 0 16
H3 Detached 17 2 15
H4 Detached 30 0 30
H5 Terrace-End 17 4 13
H6 Terrace-Mid 12 3 9
H7 Terrace-End 16 7 9
H8 Terrace-Mid 14 0 14
H9 Terrace-End 7 0 7
H10 Terrace-Mid 7 7 0
H11 Flat 13 11 2
H12 Semi-Det 14 7 7
H13 Detached 22 17 5
H14 Semi-Det 15 12 3
H15 Flat 16 7 9
H16 Semi-Det 9 9 0
H17 Detached 21 0 21
H18 Flat 7 6 1
H19 Flat 3 2 1
H20 Flat 11 8 3
Overall 311 124 (40%) 187 (60%)
Appendix A Detailed results
|
59
Table A.4 Summary of air exchange rate results
Property
details
Air exchange rate
ID PFT PFT
ADF
Table 1.1b
Measured/
ADF recommended
ach l/s l/s
P1 0.38 24.2 21.0 1.15
P2 0.34 21.4 21.0 1.02
H1 0.55 61.4 52.1 1.18
H2 0.47 26.1 24.2 1.08
H3 0.49 18.9 17.8 1.06
H4 0.71 77.5 51.7 1.50
H5 0.26 15.3 27.5 0.55
H6 0.70 35.6 21.2 1.67
H7 0.44 31.1 31.6 0.98
H8 0.62 36.7 27.1 1.35
H9 0.37 15.3 19.5 0.78
H10 0.38 15.8 19.8 0.80
H11 0.27 9.4 17.0 0.56
H12 0.99 57.2 27.6 2.07
H13 0.97 52.8 26.8 1.97
H14 0.43 21.7 23.4 0.93
H15 0.23 7.5 17.0 0.44
H16 0.20 11.4 25.1 0.45
H17 0.45 35.0 36.6 0.96
H18 0.42 14.2 17.0 0.83
H19 0.23 11.4 23.0 0.50
H20 0.24 8.1 17.0 0.47
Mean 0.46 27.6 1.01
Min 0.20 7.5 0.44
Max 0.99 77.5 2.07
60
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Table A.5 Summary of mechanical extract rates
Property
details
Extract rates
ID
Kitchen hood/
fan above
cooker
Kitchen fan
away from
cooker
Bathroom En-suite WC
l/s l/s l/s l/s l/s
P1 28.0 3.0 4.5 6.0
P2 31.0 1.5 7.0 3.0
H1 32.0 7.7 5.7 9.4
H2 28.0 6.0 15.0
H3 27.9 17.0
H4 9.0 14.9 3.4 12.1
H5 34.5 18.2 10.4
H6 32.7 4.5 3.6
H7 34.3 11.6 4.1
H8 Fan not working 15.1
H9 25.6 13.3 16.3
H10 15.2 14.3 16.6
H11 21.6 11.1
H12 34.0 10.3 3.9
H13 19.5 3.1 2.1
H14 15.1 18.5 17.8
H15 Could not test
1
15.7
H16 35.0
3
5.2 17.0
H17 25.5 5.8 6.2
H18 21.6
H19 28.8 2.2
2
7.0
H20 21.6 19.1
2
Mean 23.4 34.2 10.4 8.3 8.7
Min 9.0 32.7 1.5 3.4 2.1
Max 32.0 35.0 19.1 17.8 17.0
1
Fan fitted but unable to achieve a sufficient seal for test.
2
Difficult to seal to undertake test.
3
Tested without diffuser.
Appendix A Detailed results
|
61
Table A.6 Summary of average relative humidity levels over monitoring period
Property
details
Relative humidity
ID Outside Bathroom Kitchen
Living
room
Bedroom
% RH % RH % RH % RH % RH
H1 78.4 48.2 45.9 46.7 50.4
H2 70.1 63.4 57.0 62.2 60.3
H3 71.7 43.4 46.4 42.6 43.1
H4 60.2 43.0 39.0 36.1 39.8
H5 73.6 55.1 55.3 52.5 53.9
H6 n/a 47.6 46.9 44.4 46.8
H7 74.2 55.7 50.3 53.8 53.7
H8 73.6 63.7 53.1 53.6 58.1
H9 67.6 54.4 51.6 42.6 46.0
H10 68.8 49.1 48.1 41.2 42.5
H11 71.3 52.7 57.6 61.2 65.0
H12 67.5 57.1 52.4 52.3 56.0
H13 62.8 45.3 37.8 36.6 41.8
H14 62.6 53.7 52.7 51.8 54.5
H15 70.5 67.5 56.6 58.1 61.2
H16 n/a 66.8 60.0 66.3 62.4
H17 71.8 52.1 49.9 48.8 52.3
H18 61.0 60.6 46.5 54.1 58.7
H19 67.2 56.1 48.1 48.6 49.9
H20 69.2 71.0 53.6 58.6 71.0
Mean 69.0 55.3 50.4 50.6 53.4
Min 60.2 43.0 37.8 36.1 39.8
Max 78.4 71.0 60.0 66.3 71.0
62
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Table A.7 Summary of maximum average daily relative humidity levels
Property
details
Relative humidity
ID Bathroom Kitchen Living room Bedroom
% RH % RH % RH % RH
H1 49.7 49.4 51.0 52.3
H2 66.8 64.3 68.9 63.8
H3 45.8 49.4 46.0 45.6
H4 48.3 46.5 46.6 47.3
H5 58.8 60.2 56.1 58.9
H6 52.3 50.5 47.7 49.8
H7 58.2 52.9 55.4 55.2
H8 66.9 55.5 55.9 60.3
H9 58.3 53.9 45.0 49.3
H10 52.8 49.7 43.9 44.7
H11 61.2 59.8 63.3 66.9
H12 66.3 60.4 60.2 64.0
H13 50.2 41.2 38.6 44.9
H14 55.8 56.1 54.3 56.9
H15 73.8 60.3 60.4 62.9
H16 82.2 63.6 70.1 65.4
H17 59.2 58.8 54.2 59.2
H18 64.8 51.7 60.9 65.1
H19 58.3 50.5 51.4 51.7
H20 73.5 55.5 60.6 73.5
Mean 60.2 54.5 54.5 56.9
Min 45.8 41.2 38.6 44.7
Max 82.2 64.3 70.1 73.5
Appendix A Detailed results
|
63
Table A.8 Nitrogen dioxide results (averaged over one week)
Property details Nitrogen dioxide levels
ID House type Kitchen Outside
μg.m
-3
μg.m
-3
P1
1
Semi-Det 31.0 33.9
P2
1
Semi-Det 17.6 39.9
H1
1
Detached 15.9 21.2
H2
1
Detached 19.2 31.4
H3
1
Detached 12.9 21.1
H4
1
Detached 24.2 31.6
H5
1
Terrace-End 56.6 21.4
H6
1
Terrace-Mid 26.2 21.9
H7
1
Terrace-End 47.3 21.5
H8 Terrace-Mid 23.8 21.8
H9 Terrace-End 19.3 35.5
H10 Terrace-Mid 18.5 40.3
H11 Flat 11.4 33.3
H12
1
Semi-Det 22.9 13.2
H13
1
Detached 16.1 19.5
H14
1
Semi-Det 17.4 18.9
H15 Flat 9.6 22.9
H16
1
Semi-Det 55.0 26.1
H17
1
Detached 44.8 2.4
H18 Flat 9.6 18.1
H19 Flat 7.9 *
H20 Flat 11.3 31.1
Mean 23.6 25.1
Min 7.9 2.4
Max 56.6 40.3
* Sample missing.
1
Gas cooking.
64
|
Ventilation and Indoor Air Quality in Part F 2006 Homes
Table A.9 Total volatile organic compounds (TVOCs) results (averaged over
one week)
Property
details
TVOC levels
ID Living room
Master
bedroom
Outside
μg.m
-3
μg.m
-3
μg.m
-3
P1 53 <40 <40
P2 <40 231 <40
H1 168 224 <40
H2 285 464 103
H3 <40 156 <40
H4 112 320 <40
H5 90 111 <40
H6 114 204 <40
H7 708 976 <40
H8 219 311 <40
H9 251 330 <40
H10 191 171 <40
H11 n/a n/a n/a
H12 n/a n/a n/a
H13 n/a n/a n/a
H14 n/a n/a n/a
H15 n/a n/a n/a
H16 n/a n/a n/a
H17 n/a n/a n/a
H18 1147 1213 <40
H19 572 446 <40
H20 56 535 <40
Mean 264 379 <40
Min <40 <40 <40
Max 1147 1213 103
Appendix A Detailed results
|
65
Table A.10 Formaldehyde results (averaged over three days)
Property
details
Formaldehyde levels
ID Living room
Master
bedroom
Outside
μg.m
-3
μg.m
-3
μg.m
-3
P1 10.5 15.5 0.8
P2 15.9 19.0 6.3
H1 21.9 30.2 0.9
H2 21.2 31.0 6.7
H3 19.9 18.9 1.1
H4 12.0 24.5 2.4
H5 16.3 16.4 1.3
H6 24.0 29.0 2.0
H7 63.0 54.0 1.8
H8 14.0 22.3 3.0
H9 24.0 31.0 1.0
H10 11.7 11.4 1.8
H11 37.0 37.0 2.3
H12 19.2 19.6 2.4
H13 10.2 13.1 2.8
H14 25.2 32.0 0.9
H15 55.0 46.0 1.6
H16 46.0 36.0 1.6
H17 14.8 24.8 1.8
H18 35.0 31.0 1.5
H19 25.0 28.1 0.6
H20 27.0 50.0 2.0
Mean 24.9 28.2 2.1
Min 10.2 11.4 0.6
Max 63.0 54.0 6.7
ISBN 978-1409823759
9 781409 823759
ISBN: 978 1 4098 2375 9
Price £20

Ventilation and Indoor Air Quality in Part F 2006 Homes
BD 2702

Simon McKay and David Ross AECOM Ltd Ian Mawditt and Stuart Kirk Building Services BuildingSciences Ltd March 2010 Department for Communities and Local Government

The findings and recommendations in this report are those of the authors and do not necessarily represent the views of the Department for Communities and Local Government.

Department for Communities and Local Government Eland House Bressenden Place London SW1E 5DU Telephone: 0303 444 0000 Website: www.communities.gov.uk © Queen’s Printer and Controller of Her Majesty’s Stationery Office, 2010 Copyright in the typographical arrangement rests with the Crown. This publication, excluding logos, may be reproduced free of charge in any format or medium for research, private study or for internal circulation within an organisation. This is subject to it being reproduced accurately and not used in a misleading context. The material must be acknowledged as Crown copyright and the title of the publication specified. Any other use of the contents of this publication would require a copyright licence. Please apply for a Click-Use Licence for core material at www.opsi.gov.uk/click-use/system/online/ pLogin.asp, or by writing to the Office of Public Sector Information, Information Policy Team, Kew, Richmond, Surrey TW9 4DU e-mail: licensing@opsi.gov.uk If you require this publication in an alternative format please email alternativeformats@communities.gsi.gov.uk Communities and Local Government Publications Tel: 030 0123 1124 Fax: 030 0123 1125 Email: product@communities.gsi.gov.uk Online via the Communities and Local Government website: www.communities.gov.uk March 2010 Product Code: 09BD06323 ISBN: 978 1 4098 2375 9

5 4.3 Secondary aims Chapter 3 Study design 3.1 Introduction 2.3 4.5 Dwelling selection and recruitment Sample size Monitoring period Ventilation conditions Monitoring 5 9 10 10 10 11 12 12 13 13 14 15 18 18 18 19 19 20 20 20 21 21 Chapter 4 Methodology 4.1 3.1 Results and lessons learnt .2 3.1 4.6 4.Contents Executive summary Chapter 1 Introduction Chapter 2 Background 2.7 Air permeability Ventilation capacity Air exchange rates TVOCs HCHO Nitrogen dioxide Temperature and relative humidity Chapter 5 Pilot study 5.4 4.2 Primary aim 2.4 3.2 4.3 3.

1 6.2 6.3 7.6 Are intermittent extract flow rates sufficient? Are trickle ventilator areas sufficient? Are dwellings Part F compliant? Are dwellings Part L compliant? Should we change air pressure testing for Part L compliance? Should we undertake air pressure testing for Part F compliance? Chapter 8 Conclusions and recommendations Appendix A Detailed results 56 .6 6.3 6.4 6.7 6.5 6.4 7.5 7.Chapter 6 Results 6.2 7.1 7.9 Summary of dwellings monitored Measurements of air permeability Natural ventilation opening areas Whole house air exchange rates Intermittent extract flow rates Relative humidity Nitrogen dioxide TVOCs Formaldehyde 22 22 22 25 28 31 34 37 38 39 40 40 42 51 51 52 52 53 Chapter 7 Discussion 7.8 6.

m-2 and 30% achieve better than 5 m3. It has also investigated how well dwellings comply with Part F 2006 standards. 6. 3. relative humidity (RH) levels and other indoor pollutant concentrations. ventilation factors and indoor sources in determining concentrations of indoor pollutants.m-2 at 50 Pa. the ventilation system set up at full capacity to test whether the ventilation guidance in ADF is adequate. where possible. ADF 2006 provides guidance for new dwellings built to an air permeability down to 3 to 4 m3. 4. The results have been analysed to assess the significance of dwelling characteristics. but no additional guidance is provided. 2. whole house air exchange rates.h-1. This study was commissioned by Communities and Local Government to inform and provide evidence for the amendments to the Part F regulations and guidance documents that are due to come into force in October 2010. Since ADF 2006 was published. new evidence has emerged to suggest that there has been a significant improvement in the airtightness of new dwellings. The study has assessed whether the guidance in the 2006 edition of Approved Document F (ADF) is effective at providing adequate ventilation and good indoor air quality in new dwellings. thereby minimising the risks to health of the occupants.h-1. Diary records of occupant activities and questionnaires on the dwellings and their indoor environments were also collected. All homes were occupied with. The focus has been on more airtight dwellings which are naturally ventilated. mechanical extract flow rates. ADF accounts for the contribution made by air infiltration when determining the purpose-provided ventilation necessary to provide good indoor air quality.m-2. It suggests that additional ventilation may be required for more airtight homes where there is less air infiltration. This report presents the results from a sample of 22 occupied homes built to Parts L and F 2006 standards. . Approved Document F 2006 5. They have also been analysed to determine whether the design recommendations in ADF 2006 are being met within the dwellings.h-1. Approximately 5% of around 3000 new dwellings that have been pressure tested achieve results better than 3 m3.Executive summary | 5 Executive summary Introduction 1. The project involved carrying out measurements of airtightness.

6

|

Ventilation and Indoor Air Quality in Part F 2006 Homes

Aims
7. The primary aim of this study was to establish whether additional ventilation is required in more airtight dwellings with an air permeability equal to or better than 5 m3.h-1.m-2. The study focused on ADF ventilation system 1: “Background ventilators with intermittent extract fans” because:

ventilation in airtight properties, and in particular the reliance on natural driving forces (wind and temperature stack effects) to provide the background ventilation through trickle ventilation. permeability equal to or better than 5 m3.h-1.m-2, the analysis would be more meaningful if it focused on just one ventilation system type. 8. There were a number of secondary aims: Compliance with Part F: We wished to investigate whether the installed ventilation systems complied with Part F guidance. There is both objective and anecdotal evidence to suggest that ventilation systems are designed to comply with ADF 2006 standards but do not achieve this on construction due to inadequate installation and inspection. Air pressure testing for Part L: For Part L compliance, air pressure testing is undertaken to BS EN 13829: Method A (all mechanical ventilation systems sealed, all natural ventilation openings closed). The proposed amendments to ADL1A would change the test method to BS EN 13829: Method B (all mechanical ventilation systems sealed, all natural ventilation openings closed and sealed). We wished to investigate the impact and need for this change. Air pressure testing for Part F: We wished to investigate if an air pressure test with all natural ventilation systems fully open would provide an estimate of the equivalent area of the installed background (trickle) ventilation.

Results
9. The key results from this study are:

trickle ventilation and intermittent extract. Ten of the 22 dwellings tested (46% of total sample including pilot homes, 50% of main sample) achieved an air permeability of less than 5 m3.h-1.m-2.

Executive summary

|

7

trickle ventilation installed to meet ADF 2006 guidance. In the worst case, only just over half (57%) of the recommended trickle ventilation was installed. recommended in ADF 2006. recommended in ADF 2006. Furthermore, none of the fans located away from the cooker achieved the recommended flow rates in the kitchen. the new daily and weekly relative humidity guideline levels proposed in the consultation version of ADF 2010. However, making reasonable assumptions, it would be expected that approximately four homes would have monthly relative humidity levels that would exceed the newly proposed monthly relative humidity guideline level. recommended in ADF 2006. A key issue in these cases was insufficient installed flow where fans are located away from the cooker.

recommended in ADF 2006. It is important to note that in developing ADF 2006, TVOCs were chosen to represent organic compounds. A review of indoor air quality results suggested that if the TVOC guideline level of 300 μg.m-3 was met, then individual chemical health-based guideline levels (eg benzene, toluene, formaldehyde) should also be met.

Conclusions
10. The key conclusions are:

recommended natural ventilation provisions in ADF 2006 are sufficient for airtight homes. The study is relatively small and the conclusions should only be treated as indicative. Furthermore, in all cases, the capacity of the ventilation system did not meet that recommended in ADF 2006. that they are installed correctly to provide the capacity recommended in ADF 2006. There is no evidence from this study that with correct installation, the extract rates should be increased for more airtight dwellings. issues for the ventilation system as a whole, at least a significant minority of TVOC levels would be expected to exceed the recommended guideline level,

8

|

Ventilation and Indoor Air Quality in Part F 2006 Homes

particularly for the more airtight homes. VOCs are produced by building products and the activities of the occupants such as smoking, use of personal hygiene products and interior decorating. It may seem pedantic to change the design of a ventilation system just to better control one pollutant, but the long term effects on health due to exposure to VOCs are not well documented, and it seems prudent to err on the side of caution. Furthermore, the TVOC criterion was selected as a sensitive marker for individual organic chemical compounds (ie if TVOC levels are below the criterion, previous research suggest that individual organic chemical compounds would also be below recognised indoor or outdoor health-based levels for these pollutants). There are also other hazards to health that are affected by ventilation, such as house dust mites, which are associated with respiratory illnesses including asthma. We do not currently attempt to control these under Part F because their breeding success is influenced by heating and hygiene practice as much as ventilation. Improving ventilation would be step in the right direction to limit this risk. the flats had two façades, and were ventilated as ‘multi-sided façades’ according to ADF 2006, they were effectively single façades as the second façade was limited. The ventilation and indoor air quality results suggest that these homes were under-ventilated. system to full capacity (trickle vents are not always open and fans are not always used). Hence, the actual pollutant levels in dwellings will likely be higher than those recorded in this study.

Recommendations
11. The key recommendations are:

inspected to provide the ventilation capacity as designed. study. However, there are sufficient grounds to support the proposal to increase trickle ventilators in dwellings having an air permeability equal to or tighter than 4 m3.h-1.m-2. This is based on the levels of TVOCs observed, and the implications for individual organic compounds. This could also have benefits in controlling other hazards to health, but this is difficult to quantify. findings of this study, to better determine indoor air quality levels achieved in airtight dwellings in which the ventilation system is installed correctly. reducing source strength by product controls.

ventilation factors and indoor sources in determining concentrations of indoor pollutants. A key secondary aim was to investigate how well dwellings complied with Part F 2006. This report presents the results from a sample of 22 occupied homes built to Parts L and F 2006. Analysis of the results was used to assess the significance of dwelling characteristics. relative humidity (RH) levels and other indoor pollutant concentrations. the ventilation system set up at full capacity to test whether the ventilation guidance in ADF is adequate. where possible. mechanical extract flow rates. whole house air exchange rates. All homes were occupied with.Chapter 1 Introduction | 9 Chapter 1 Introduction This study was commissioned by Communities and Local Government (CLG) to assess whether the guidance in the 2006 revision of Approved Document F (ADF) is effective at providing adequate ventilation and good indoor air quality in new dwellings. It is intended that the results from this study will help inform the 2010 amendments to Parts F and L of the Building Regulations. Diary records of occupant activities and questionnaires on the dwellings and their indoor environments were also collected. The results were also analysed to determine whether the design recommendations in ADF 2006 were being met within the dwellings. The study particularly focused on more airtight dwellings which are naturally ventilated. thereby minimising the risks to health of the occupants. . The project involved carrying out measurements of airtightness.

due to less air infiltration.10 | Ventilation and Indoor Air Quality in Part F 2006 Homes Chapter 2 Background 2. In the main. It accounts for the contribution of air infiltration in determining the purpose-provided ventilation necessary.h-1. Hence additional guidance is proposed in the consultation version of ADF 2010 for the more airtight homes.h-1. it reviews the necessity of increasing ventilation provisions for more airtight dwellings. no additional guidance was provided for these dwellings at that time. Approved Documents are intended to provide guidance for some of the more common building situations.h-1. Approximately 5% of new dwellings pressure tested achieve results better than 3 m3. Since ADF 2006 was developed. This is based on a total sample of approximately 3000 new dwellings and arises from two sources which agree with one another: a confidential industry analysis and Building Sciences Ltd (BSL) air pressure testing results. As it was expected that in the near future relatively few dwellings would approach or be tighter than this level of air permeability. However. . new evidence suggests that there has been a significant improvement in the airtightness of new dwellings.m-2 and 30% achieve better than 5 m3. The primary aim of this study is to provide evidence to determine whether additional ventilation provisions are required for those more airtight dwellings. It suggests that additional ventilation provisions may be required for more airtight homes. the ventilation system specifications are not amended for dwellings designed to an air permeability leakier than 5 m3.m-2.2 Primary aim ADF 2006 provides guidance for new dwellings built down to an air permeability of 3–4 m3.m-2.m-2 at 50 Pa.1 Introduction This study was undertaken to provide evidence for the proposed amendments to Part F 2010. It also addresses other issues such as compliance with Part F. In particular. increased ventilation provisions are recommended for dwellings designed to be tighter than this.h-1. 2.

This would be undertaken similarly to the testing above but with all natural ventilation systems fully open. Air pressure testing for Part F: We wished to investigate the use of a further air pressure test for Part F. There are three key reasons for this focus: ventilation in airtight properties.m-2 as described later. The consultation version of ADL1A proposes changing the test method to BS EN 13829: Method B (all mechanical ventilation systems sealed. with an air permeability equal to or better than 5 m3. 2. all natural ventilation openings closed). it particularly focuses on ADF ventilation system 1: background ventilators with intermittent extract fans.3 Secondary aims There were a number of secondary aims to this study.h-1.h-1.m-2. Furthermore. We wished to investigate the impact and need for this change. the reliance on natural driving forces (wind and temperature stack effects) to provide the background ventilation through trickle ventilation. . better analysis could be undertaken by focusing on just one ventilation system type – with the expectation that inferences could be made to other system types. air pressure testing is undertaken as per BS EN 13829: Method A (all mechanical ventilation systems sealed. Review compliance with Part F: We wished to investigate whether the installed ventilation systems complied with Part F. In particular. all natural ventilation openings closed and sealed). The proposal was that by comparing these results with those of Method B (with natural ventilation systems fully sealed). There was a mix of objective and anecdotal evidence to suggest that the ventilation systems were designed to ADF 2006 but did not achieve this on construction due to inadequate installation and inspection. Air pressure testing for Part L: For Part L compliance.Chapter 2 Background | 11 with air permeability equal to or better than 5 m3. it could provide an estimate of the equivalent area of background (trickle) ventilation installed in dwellings to confirm installed capacity met design intentions.

h-1. are located. we modified the criteria to monitor any dwelling constructed to Part F/L 2006 (air permeability should be better than 10 m3. the building drying-out. To meet the primary aim of the study.m-2 high indoor pollutant emissions from.1 Dwelling selection and recruitment The initial selection criteria for dwellings were as follows. for example.h-1.h-1.m-2). contact was with the householder. new construction materials and furnishings. This in itself is a valuable study as we are assessing the performance of newly introduced regulations. No specific occupancy levels were selected – although we looked to obtain the sample from both private and social housing to provide a mix. in a majority of the cases they did not have details of the airtightness of their dwelling to determine if it was equal to or better than 5 m3. Many dwellings constructed since 2006 had planning permission to pre-2006 Building Regulations. As a result.m-2.12 | Ventilation and Indoor Air Quality in Part F 2006 Homes Chapter 3 Study design 3. we would aim to gain data on the performance in the most airtight homes through: . 3 . and lived in for six months. and painting and decorating. who were carrying out the testing. they could give in finding these homes given the current economic climate. Limited. These initial criteria proved challenging for three principal reasons.

We decided to focus only on ADF ventilation system 1 (trickle ventilation with intermittent extract). to confirm the techniques for measurements and to check the suitability of the household questionnaires and diaries that had been developed for the study.h-1. The sample was obtained through a combination of the following routes: A disturbance payment of £100 was paid to the householder to take part in the study.m-2 the trend in performance from leakier to more airtight dwellings. While the total sample number was relatively small.h-1. The aim of the pilot study was to assess the protocol developed for this project in terms of practicality. 3. reliability and variability.2 Sample size An initial pilot study of two dwellings was undertaken. The sample size for the main study was 20 dwellings with the intention of at least half having an air permeability of better than 5 m3. Additional reasons for focusing on this ventilation system were discussed in Section 2. it was considered that it should give a good indication if issues of poor indoor air quality or poor compliance with Part F were common.m-2.3 Monitoring period The intention was to undertake the main sampling in March and April 2009. Information about the nature of the testing was provided in advance.Chapter 3 Study design | 13 permeability better than or equal to 5 m3. . This should allow us to better determine the trend line in performance as airtightness varies (ie removes variation due to different ventilation system types). 3.

etc. In practice. It was thought sufficient to provide the information required for that dwelling.14 | Ventilation and Indoor Air Quality in Part F 2006 Homes period. However.) ‘closed-up’. Hence the following steps were undertaken: occupants asked to keep them open capacity during all bathing occasions during cooking times A daily diary was given to the occupants to complete during the monitoring period to confirm that the above instructions were followed and. relatively little use of windows. if not. this methodology provides the reasonably worst case internal RH levels. . 3.4 Ventilation conditions To assess whether ADF 2006 recommendations were adequate. this aim was broadly met with the main study taking place between March and mid-May. which would be lowest during the coldest winter periods. The initial pilot study took place in January/February 2009. as highlighted below. it was necessary for the ventilation system to be used to its full capacity.e. the study made requirements of the occupants and based on experience it was considered that the necessary occupant behaviour may not have been reliably met over longer periods. Furthermore. high and there is less warming of the outside air (and reduction in RH) as the air enters the home. (It is noted that the actual risk of condensation should also take account of the internal surface temperatures. the difference between internal and external temperatures would be reduced and thus the stack driving forces lower. why they needed to make changes. Each dwelling was monitored for one week. i.

published by the Air Tightness Testing and Measurement Association (ATTMA). 1 Technical Standard 1 (TS1). 3.5 3. Comparison of the latter two tests provides a means of estimating the equivalent area of trickle ventilation in the dwelling to compare against what appears to have been installed.1 Monitoring Introduction The strategy of the study was based on previous research experience of measuring airtightness. the method referenced for Building Regulations compliance is Method A. Comparison of the first two tests will allow an evaluation of the advantage of including Method B instead of Method A1 as the means for compliance testing for dwellings. Method B is specifically amended in TS1 to remove the requirement to temporarily seal trickle ventilators. for the purpose of this report. A summary of the parameters measured in this study is provided in Table 1.5. 3. currently refers to BS EN 13829 Method B as the compliance testing method.3 Ventilation The perfluorocarbon (PFT) technique was used to provide an estimate of the ventilation rate in the dwelling over the one week sampling period. Hence. 3.2 Airtightness The following airtightness tests were carried out on the first visit to each dwelling: ventilation openings closed) ventilation openings closed and sealed) ventilation openings opened. However.5.4 Extract flow rates A rotating vane anemometer with an aircone hood attachment was used to determine the air flow rate of each extract fan and cooker hood in the dwellings at its highest setting.5. quality assured data. .5. validated techniques and methods of investigation in order to provide reliable. The parameters selected are further discussed below and the next section provides more detail of the individual measurement techniques. ventilation and indoor air quality in homes and used as far as possible well proven.Chapter 3 Study design | 15 3.

During the initial visit to the property.5.5. are covered by performance criteria in ADF Appendix A. use of extract fans and window opening. Formaldehyde was also measured as there have been suggestions to introduce formaldehyde into the Appendix A performance standards.6 Questionnaires/diaries Each occupant who participated in the study answered a questionnaire concerning his/her home. without being biased from later stages of the study. In addition. the occupant completed a diary regarding gas cooking activity. . with the exception of formaldehyde. after we had informed them of the appropriate utilisation of the ventilation system and reasons for doing so. the research team recorded: the top of the floor finish) WCs This was undertaken to provide information about how the occupants would normally operate their ventilation system.5 Indoor air quality The following indoor pollutants were monitored: 2 ) All of these pollutants.16 | Ventilation and Indoor Air Quality in Part F 2006 Homes 3. 3. their normal ventilation behaviour and the activities in the home during the sampling period.

WC and kitchen) Main study 2 March to 15 May 2009 20 NO2 (kitchen and outside) TVOCs (living room.Chapter 3 Study design | 17 Table 1 Summary of the IAQ and ventilation monitoring during the pilot and main study Pilot study Sampling period No. master bedroom and outside) Air permeability measurements (three types) PFT technique for whole house ventilation Trickle ventilator area noted (not measured) in each room Internal door undercuts Mechanical extract fan flow rates (bathroom. and outside) HCHO (living room. ensuites. master bedroom. master bedroom and outside) Temperature and humidity (living room. bathroom and outside) HCHO (living room. master bedroom and outside) Temperature and humidity (living room. of homes IAQ parameters and locations monitored 29 Jan to 5 Feb 2009 2 NO2 (kitchen and outside) TVOCs (living room. master bedroom and outside) Air permeability measurements (three types) PFT technique for whole house ventilation Internal door undercuts Mechanical extract fan flow rates (bathroom. WC and kitchen) Ventilation rate measurements . ensuites. master bedroom. kitchen.

ie six tests per dwelling. Each of the above test methods was performed using pressurisation and depressurisation procedures. ie 10 mm clear gap. The final air permeability results for each method are determined as the average of the pressurisation and depressurisation results with an overall uncertainty of less than ±10%.2. Door undercuts to internal doors were measured to confirm adequacy of cross ventilation provision. or .2 4. using UKAS calibrated domestic standard ‘blower door’ fan and micromanometers and associated equipment.18 | Ventilation and Indoor Air Quality in Part F 2006 Homes Chapter 4 Methodology 4. the equivalent areas were recorded.2 Mechanical The volumetric flow through all intermittent extract fans and cooker hoods was measured within each dwelling using a UKAS calibrated Airflow AV-2 rotating vane anemometer with an aircone hood attachment to encompass the fan.2. a geometric area measurement of the trickle ventilator was made.1 Ventilation capacity Natural Approved Document F (ADF) recommends that the equivalent area is marked on the trickle ventilator in any easily visible location where practical. Where available. In the few cases where this information was not visible. 4. The total equivalent area for each dwelling and associated dwelling floor areas were referenced to the recommended equivalent areas published in Table 1.2a of ADF. 4. openings closed and sealed.1 Air permeability The dwellings were air permeability tested using the following three methodologies: natural ventilation openings closed.

. and passively collected on adsorption tubes.g. The overall uncertainty with this measurement technique for the flow rates measured is less than ±5%. The PFT technique is based on the fact that the average air infiltration rate is approximately equal to the reciprocal of the time averaged indoor tracer gas concentration and. Following the sampling. The PFT sources and samplers were obtained from PentIAQ. After the sampling period. and the laboratory analysis. site limitations and number of sources/samplers.3 Air exchange rates The PFT technique is a passive sampling technique for measuring ventilation rates in buildings. The total number of sources in each room (source strength) was arranged to give an approximately uniform source rate per unit volume throughout the house. the time averaged indoor tracer gas concentration is determined. a perfluorocarbon compound is passively emitted from small tracer sources. PFT sources were set out in all rooms in each house. which combined. pre-conditioned with Tenax® TA at the analytical laboratory (Scientifics). but not in small or wet rooms (e. The concentration of total volatile organic compounds (TVOCs) was determined by the total ion current response of the individual compounds. 4. Chemical compounds were then identified via a mass spectrometric (MS) detection library. they were then returned to PentIAQ for analysis. The samplers were placed in the living rooms. the sampling rate. The readings were all taken with the fans and hoods at their highest flow setting over a 30 second averaging time. living room and garden. store cupboards and bathrooms). using this technique. including hallways and stairwells.Chapter 4 Methodology | 19 ceiling/wall terminal. were installed in each house for a period of seven days. The overall uncertainty of this technique is a combination of the source emission rate. Three ATD tubes were installed at each house: the master bedroom. 4. The results for the effective air exchange rate for the houses are derived from the emission rate per volume (room) and the local mean age of air. the tubes were analysed at the laboratory by automatic thermal desorption–gas chromatography (ATD–GC).4 TVOCs Passive automatic thermal desorption (ATD) tubes. In this technique. kitchen and all bedrooms. quantified using the calibrated response factor of toluene. Sweden. The amount of tracer adsorbed depends mainly on the emission rate from the source tubes and the dilution of it by ventilation air. with a known constant emission rate. which is described in ISO-standard 16000-8. provides an accuracy of ± 10%.

respectively. which contains a mesh treated with 50% triethanolamine (TEA) and 50% acetone that adsorbs NO2. The loggers were set to record at 5-minute intervals during the 7-day monitoring period and were generally placed between 1000–1500 mm above floor level to minimise variations induced by thermal stratification. The HCHO badges were supplied and analysed by the analytical laboratory (Scientifics). This method relies on the transfer of NO2 by diffusion to a collector at one end of the tube.5 HCHO Passive formaldehyde (HCHO) badges were installed for three days consecutive sampling using SKC UNME00 samplers using 2. The Palmes tubes were supplied and analysed by the analytical laboratory (Scientifics) and were installed in the kitchen and garden of each property. At the end of the monitoring. 4. at least 1 metre from vertical surfaces.4dinitrophenyl-hydrazone (DNPH) as the adsorbent packing material.3% and ±2% for temperature and relative humidity.0 μg. Limit of detection (LOD) was 1. living room and garden of each house. main bathroom and garden of each property. At the end of the sampling period. and were installed in the master bedroom. The typical accuracy of the loggers used was ±0.20 | Ventilation and Indoor Air Quality in Part F 2006 Homes 4. The concentration of HCHO in the air was determined by subtraction of HCHO adsorbed onto the blank (correction) tape from that adsorbed onto the exposed sample tape.6 Nitrogen dioxide The average mean concentration of NO2 was measured using passive diffusion (Palmes) tubes over the 7-day sampling period. the mean NO2 concentration was quantified using a segmented flow autoanalyser with ultraviolet detection. The reported accuracy is ±10%. kitchen.m-3 with an accuracy of ±15%. .7 Temperature and relative humidity Hygrothermal conditions were recorded using combined temperature and relative humidity USB data loggers located in the master bedroom. Desorption and analysis of the diffuse samplers was performed using a diffuse high Performance liquid chromatography (HPLC) using UV absorption between 230 and 370 nm. the data were downloaded for later analysis. 4. living room.

in particular the criteria for recruitment. This resulted in modifications to the study design. measurement techniques and questionnaires and diaries. everything worked as intended and as a result we have included the pilot study results with the main study results (see Section 7). RH was not measured in the kitchen and the bathroom in the pilot homes but was in the main study. Overall. The main lesson learnt was the difficulty in obtaining the sample group (discussed previously in Sections 2 and 3).1 Results and lessons learnt The aim of the pilot study was to assess the protocol.Chapter 5 Pilot study | 21 Chapter 5 Pilot study 5. . the pilot study only recorded the number of trickle vents whereas in the main study we also recorded the equivalent area. In addition.

the . 6.m-2.1 Summary of dwellings monitored The sample of 22 homes (including the pilot study). comprised the following dwelling types all built to Parts F and L 2006. The individual test results for each dwelling are presented in Table A. 50% of main sample) achieved an air permeability of less than 5 m3. Figure 2 shows how the air permeability varies by dwelling type (P1–P2 are the pilot homes and H1–H20 are the main study homes).7 m3.h-1. air pressure test Method A is the current test method used for Part L compliance purposes. All mechanical ventilation systems are sealed and all natural ventilation openings are closed but not sealed. 6.h-1. the flats were the most airtight dwelling type in this sample – all achieving an air permeability of less than 5 m3. Note that while we looked to bias this sample towards more airtight dwellings.m-2.m-2.2 6.22 | Ventilation and Indoor Air Quality in Part F 2006 Homes Chapter 6 Results 6. with a mean of 3.9 m3.h-1.h-1.2. This is discussed further in Section 7. Each dwelling type had at least one dwelling with an air permeability of better than 5 m3.h-1. Figure 1 shows the distribution of the air permeability for each dwelling. The mean air permeability of the remainder of the dwellings (houses) was 6.2. Ten of the 22 dwellings tested (46% of total sample including pilot homes.1 Measurements of air permeability Introduction This section provides a summary of the results of the air permeability tests undertaken at all 22 homes.m-2 required by ADL 2006 for dwellings sampled. Overall.2 Air pressure testing with trickle vents closed (Method A) As referred to in Section 4.m-2. Two homes exceeded the maximum level of 10 m3.m-2.1.1 of the Appendix.h-1.

m-2 is similar to the mean value of approximately 6.h-1.Chapter 6 Results | 23 overall mean value of 6.h-1.m-2 obtained from a sample of approximately 3000 new dwellings combined from two sources: a confidential industry analysis and Building Sciences Ltd (BSL) unpublished air pressure testing results (from their air pressure testing services). Figure 1 Air permeability distribution (Method A) Figure 2 Air permeability distribution by dwelling type .2 m3.5 m3.

3% or 0. Figure 3 Reduction in air permeability with vents sealed 6.h-1. air pressure test Method B was similar to Method A with the addition of all natural ventilation openings (in this case trickle ventilation) being sealed instead of simply closed. we would have expected a value perhaps twice this value.h-1. .4 Air pressure testing with trickle vents open As referred to in Section 4.2.m-2 @ 50 Pa. They show an increase of only 2. The results show a smaller than expected difference compared to the results with trickle vents sealed (Method B).24 | Ventilation and Indoor Air Quality in Part F 2006 Homes 6. This test method is proposed in the consultation version of Part L 2010 to replace Method A. The exception is that all natural ventilation openings (in this case trickle ventilation) are fully open.1. The relative merits of Methods A and B are discussed in Section 7. In the development of ADF 2006.m-2 @ 50 Pa on average on fully opening the trickle ventilators.2.1.6 m3. The mean reduction was 3.3 Air pressure testing with trickle vents sealed (Method B) As referred to in Section 4. The results are shown in Figure 4. these tests were similar to Methods A and B.2 m3. The results show a small reduction in air permeability with trickle ventilation sealed. Causes for these low readings are discussed further in this and the next section. Figure 3 shows this reduction in air permeability. and using the 1/20th rule-of-thumb conversion between air permeability and infiltration.

For results of individual dwellings see Table A.3 6.1 Natural ventilation opening areas Introduction This section provides a summary of the results of the ventilation provisions installed. However.2a.Chapter 6 Results | 25 Figure 4 Increase in air permeability with trickle ventilators open 6. Nearly three-quarters of the dwellings (72% of sample) did not have sufficient trickle ventilation installed to meet ADF (2006). 6.3. the maximum provision of trickle ventilation was 162% of that required in a home with a high window to façade ratio.3. At worst.2 Recorded trickle ventilation equivalent areas Figure 5 shows the provision of trickle ventilation in the dwellings compared to that recommended in ADF (2006) Table 1. just over half (57%) of the recommended trickle ventilation was installed. data were not recorded in the pilot study and only 18 of the 20 homes in the main study had the equivalent area visible on the trickle ventilator and are reported here.2). Further review of the data suggests that the provision of trickle ventilation in many homes was based on one trickle ventilator per window installed rather than reviewing the overall requirement of the dwelling. This under-provision of trickle ventilation will partially explain the lower than expected readings for air pressure test Method C (see Section 6. In contrast. This may be insufficient. this is not expected to be the full explanation.2 in the Appendix. The results from the air pressure tests . Of the sample of 22 dwellings. depending on the size of trickle ventilator installed.

For the purpose of this study. surroundings to the actual installation and this may not be the case. This was to assess whether it was possible to measure simply on-site whether the equivalent area written on the trickle ventilator area was accurate. using Methods B and C pressure testing results.(P1/P50)n. Three steps were undertaken in analysis: Step 1: The total equivalent area at 1 Pa was determined for each dwelling using Method B (trickle vents sealed) and Method C (trickle vents open). It is also worth noting that only 40% of the trickle ventilators were open when the monitoring team arrived at the property (see Table A.2a requirement 57% 6.(r/2P1)0.2).5 .3.3 in the appendix). Figure 5 Installed trickle ventilation versus ADF Table 1.3 Measured trickle ventilation equivalent areas An alternative method was used to determine the equivalent area.(1/Cd). Two sources of errors in the laboratory testing were considered possible. The area of trickle ventilation recorded provides a possible explanation for 6% (ie one-tenth) of this.Q50.26 | Ventilation and Indoor Air Quality in Part F 2006 Homes suggest that we are only installing 50% of the required trickle ventilator area on average. using the following equation: A=1000. the trickle ventilators were all left in the open position. This may provide a further explanation of the lower than expected readings for air pressure test Method C (see Section 6.

the flow rate exponent (‘n’) is used to extrapolate down to 1 Pa. The results were generally smaller than the total equivalent areas recorded on the trickle ventilators. The equivalent area reported on a trickle ventilator is also the equivalent area at 1 Pa. ADF (2006) recommends that doors have a minimum undercut of 10 mm above the floor finish.3.61 r = air density (kg/m2). Within the equation. taken as 1. When internal doors are closed. P50 = the pressure across the vent at 1 and 50 Pa n = flow rate exponent for the air pressure test. there needs to be sufficient gap around the doors to maintain adequate air flow. However. Hence. the equivalent area was calculated using ‘n’ from pressurisation and ‘n’ from depressurisation fan tests and an average taken. The evaluation of ‘n’ is not sufficiently accurate to do so in a robust manner. less contribution to cross ventilation within the house). these rooms may well be under-ventilated when their door is closed as well as the ventilation for the whole house being lower as the contribution of the trickle ventilation in that room to the whole house ventilation will be reduced (e. taken as 0. Step 2: For each of the two methods. Of the 127 doors measured.Chapter 6 Results | 27 where: A = the background ventilator equivalent area (mm2) Q50 = the air supply rate at 50 Pa (l/s) Cd = the discharge coefficient. This should give the equivalent area for the trickle ventilators only at 1 Pa. 6. it is considered that the results from this analysis are not sufficiently reliable. It would be necessary to undertake fan flow rate tests at lower pressures to improve this accuracy. The air pressure testing results were at 35 Pa and above. Figure 6 summarises the door undercut in the sampled dwellings. Step 3: The equivalent area from Method B was subtracted from Method C.g. To achieve this.2 P1. 52% of all doors failed to achieve this 10 mm gap. . While some doors had larger gaps this will not make up for smaller gaps – the increase in air flow as the gap becomes larger follows a law of diminishing returns.4 Door undercuts Natural ventilation relies on air flow between rooms – cross ventilation across the dwelling and/or stack ventilation up through the dwelling.

6 Master Other Bathroom En-suite WC bedroom bedrooms 20 12. For results of individual dwellings see Table A.2 1 10 3.7 1 21 5.4 0 20 5.4 17 7.6 33 11. it appears that the bedrooms have the largest undercuts. All others had at least one room with a gap of less than 10 mm.5 6 8.4 in the Appendix.4.2 4 20 4.28 | Ventilation and Indoor Air Quality in Part F 2006 Homes Table 2 shows that there is some variation in door undercut depending on the room.3 2 22 4.4 6.1 Whole house air exchange rates Introduction This section provides a summary of the results of the air exchange rates. .0 5 13 2. Only one dwelling had at least a 10 mm gap for all rooms measured.6 2 15 4.8 13 8. Within this relatively small sample. It is unclear why this may be the case and how representative this is of the building stock.6 18 7. Figure 6 Size of door undercuts Table 2 Distribution of door undercuts for different rooms Lounge Kitchen Number Mean (mm) Min (mm) Max (mm) St Dev (mm) 19 6.5 6.

Figure 8 shows the air exchange rates for both flats and other dwellings.4. This variation depends on the natural ventilation driving forces and occupant ventilation behavior. in fact. Overall. window opening may be required. effectively they were single-sided properties and ADF 2006 recommends additional ventilation for such properties. The trickle ventilator areas are sized in ADF 2006 for the winter period. as shown in Section 6. Further checking is also required as to how well this marked equivalent area (based on laboratory tests) represents the actual equivalent area in practice.28 ach whereas it was 0.51 ach for houses. Half of the door undercuts are below the recommended size. Figure 9 compares the ventilation rate monitored for each dwelling against the background ventilation rates recommended in ADF Table 1. only 94% of the required area is installed. v. with the former for two-storey dwellings and the latter for flats. . Many of the flats were classed as multi-sided ventilation but.2 PFT measurements Figure 7 shows the distribution of air exchange rates measured in the dwellings using the PFT technique. Hence. iv. A number of factors have been identified to help explain these low results: i. thereby lowering the natural ventilation driving forces through the trickle ventilators. The mean air exchange rate for the flats was 0. This analysis takes into account that the recommended background ventilation rate depends on the size of the dwelling and its intended occupancy.Chapter 6 Results | 29 6.1b. The PFT approximates the air exchange rate. the approach will tend to underread the true air exchange rate. This will have particular significance if the occupant closes their internal doors. based on the equivalent area marked on the trickle ventilators. None of the flats achieved the ADF recommended background ventilation rates while 60% of the other dwellings did. there is under-installation of trickle ventilator area in these dwellings. ii. most of the ventilation was on one side of the dwelling and limited ventilation along a second side. If the ventilation rate varies much during the measurement period. iii. Based on some computational modeling work undertaken for this study. On average. ADF highlights that due to lower temperature differences between inside and outside during the spring and autumn periods. it is estimated that the air exchange rates may have under-read by between 5–15%.3. The results suggest that there should be greater variation in ventilation rates for flats as they are reliant on just wind for ventilation whereas for two-storey dwellings they are reliant on both wind and stack forces which smoothes out variations in the ventilation rate.

Figure 7 Air exchange rate estimated using the PFT technique Figure 8 Air exchange rate for individual dwellings .30 | Ventilation and Indoor Air Quality in Part F 2006 Homes This is further discussed in the next section when the results are all considered together.

5 of the Appendix. six of the 13 dwellings passed. 6.5.5 6. with two achieving the ADF recommended flow rate of 30 l/s. the fan had been replaced with a quieter fan at the occupants’ request. The following is a summary of the findings. . The mean extract flow rate for these fans was 23 l/s.1 Intermittent extract flow rates Introduction Mechanical extract flow rates were measured wherever practically possible.2 Kitchen mechanical extract Figure 10 shows the distribution of the measured flow rates for the 13 dwellings monitored with an extract over the cooker (all exhausted to the outside). In this case. Allowing for a 5% error in the measurement technique (ie pass rate at 28 l/s). The quieter model is likely to have had a lower fan speed and therefore flow rate. A complete set of results is provided in Table A.5. The minimum flow rate recorded was 9 l/s.Chapter 6 Results | 31 Figure 9 Air exchange rate estimated using the PFT technique vs ADF 2006 6.

3 Bathroom mechanical extract Figure 12 shows the distribution of the measured extract rates in the main bathrooms of 21 dwellings (not practical to measure the remaining dwelling) .32 | Ventilation and Indoor Air Quality in Part F 2006 Homes Figure 10 Kitchen extract above the cooker Figure 11 shows the distribution of the measured flow rates for the five dwellings monitored with an extract that was not directly above the cooker (all exhausted to the outside). thus significantly below the value recommended in ADF. Figure 11 Kitchen extract not above the cooker 6. All five fans were measured with a flow rate between 33–35 l/s.5. ADF recommends that these fans have a higher flow rate of 60 l/s as they are less effective in removing cooking pollutants (fans above or close to the cooker can remove some cooking pollutants before they become mixed with the room air). It is possible that they were installed to meet the previous 30 l/s criterion.

4 WC mechanical extract Figure 13 shows the distribution of the measured flow rates for all 12 dwellings that had a small WC with an extract fan. ADF recommends that extract rate should achieve 6 l/s. Figure 12 Bathroom extract rates 6. Only eight (just over one-quarter) of the sample met this criterion.5.Chapter 6 Results | 33 and en-suite bathrooms of eight dwellings. Seven of the fans achieved this criterion. Allowing for a 5% experimental error (pass rate of 14 l/s). ADF recommends that extract rates should achieve 15 l/s. . allowed a further two fans (10 in total) to pass. There appeared to be no difference between the performance of main and en-suite bathroom fans.

Problems noted included unnecessarily long and flexible duct runs and a high number of bends. kitchen. One possible reason is that the incorrect fan rating may have been installed. bathroom and master bedroom for the main sample (RH measurements were not taken in the kitchen and bathroom of pilot study homes). On several occasions. Another potential cause is inadequate installation. The consultation version of ADF 2010 proposes new RH criteria to be met during the heating season in each room: . 6.5 Discussion A significant number of mechanical extract fans were not providing the recommended flow rates. where the householder had sufficient time and there was easy access. the ducting installation was viewed.6 Relative humidity This section provides a summary of the relative humidity (RH) results measured in living room.5.34 | Ventilation and Indoor Air Quality in Part F 2006 Homes Figure 13 Small WC extract rates 6. The study was not focused on investigating installation issues.6 and Table A. For results of individual dwellings see Table A.7 in the Appendix.

The following steps were undertaken: i. The London TRY (Test Reference Year) data set was used for external climate conditions. The results showed four homes with weekly average levels above 70%. H16 had the lowest air exchange rate recorded in this study (0. We determined the maximum weekly vapour pressure level over the heating season (SAP 2005 runs suggested that daily average external temperature should be below 10. H16. the results for each room were determined and the maximum chosen. iii.Chapter 6 Results | 35 Figure 14 shows the distribution of the maximum average weekly RH level monitored in each dwelling. The average weekly vapour pressure level was determined for each room in each dwelling over the week monitoring period from the measured temperature and RH levels. If necessary. The worst case weekly vapour pressure level from (iii) was added to the excess vapour pressure level from (ii). H15. none of the flats achieved the ADF recommended background ventilation rate. This was reconverted back to RH level for each room. cooking and bathing) as usual. Some additional analysis was undertaken to indicate the likely impact to the internal RH levels from reasonably worst external RH levels. This was similarly undertaken for the external vapour pressure level. Within each dwelling. and as previously noted in Section 6. This is because internal levels are very dependent on external levels. Three of these dwellings are flats (H11.2 ach). iv. The vapour pressure excess was determined for each room by subtracting the vapour pressure outside from the vapour pressure in the room.7%).4. The reasonably worst case vapour pressure level for outside was determined. ii. If it is assumed that the average weekly level is representative of the average monthly level. The maximum average weekly level recorded was 71% which is within the new criterion of 75%.g. but only one (H8) exceeded 75% (achieved 75. v. H20). It seems reasonable that for these four homes the weekly levels would be similar to the monthly level. then four homes recorded RH levels of 65% or greater (H11. The maximum average daily level recorded was 82% which is within the new criterion of 85%. as the external levels were close to the average in the sample and the occupants were asked to undertake moisture generating activities (e. H15 and H20). Figure 15 shows the distribution of the maximum average daily RH level monitored in each dwelling.5 °C during the heating season). windows can be opened .

Further analysis could usefully investigate the impact of external moisture levels on the daily average and whether the TRY data-set is the most appropriate one to use for such analysis.36 | Ventilation and Indoor Air Quality in Part F 2006 Homes to remove excess moisture due to periods of high external levels. Figure 14 Highest weekly average RH level in each dwelling Figure 15 Highest daily average RH level in each dwelling .

This increased to 29 μg. this is a plausible explanation for the high NO2 levels. For results of individual dwellings see Table A.7 Nitrogen dioxide This section provides a summary of the results of the nitrogen dioxide (NO2) levels. The occupant commented in the daily diary that the noise level of the fan was a nuisance. therefore the fan was not used. Hence. and achieved approximately 50% of the recommended flow rate. Four dwellings exceeded the recommendation in ADF that NO2 levels should not exceed 40 μg.8 in the Appendix. Figure 16 provides the distribution of the average NO2 levels measured in the kitchen over the monitoring period.m-3 as a long term average. cooking even though requested to do so. The mean concentration of NO2 recorded in the sample was 24 μg.Chapter 6 Results | 37 6.m-3.m-3 if you consider only those 64% of dwellings that used gas for cooking. Figure 16 Distribution of NO2 concentrations .

For one of the dwellings above 600 μg. Figure 17 Highest TVOC concentration in each dwelling .38 | Ventilation and Indoor Air Quality in Part F 2006 Homes 6. overall there is no clear indication of a particular source or sources of TVOCs that is causing this high level. H8) and the results did not show this significant rise.m-3. even if just below 300 μg. Given the accuracy of the detection technique. these levels could actually have been below 300 μg. 53% were higher than the guideline figure of 300 μg.m-3 (or even higher). There was an error in the analysis of a batch of TVOC sample tubes which voided results for seven of the dwellings. they are approaching this guideline value and hence still a cause for concern. there was no obvious source from the completed questionnaire and the occupants had already moved by the time we followed up after the monitoring period. Three of these dwellings had levels between 300–330 μg. For the other dwelling above 600 μg.m-3. For results of individual dwellings see Table A. Hence it is not clear whether smoking is the cause. H6. smoking also took place in three other dwellings (H5. However. the occupant did smoke in the main bedroom (the room of the highest reading) which may have significantly impacted on the results directly or indirectly through the use of any odour-masking chemical products.m-3 (H7).m-3 in ADF 2006. Reviewing the data.m-3.8 TVOCs This section provides a summary of the results of the TVOC levels measured in the living room and master bedroom.9 in the Appendix. However. Figure 17 provides the distribution of the highest TVOC levels monitored in each home.

although somewhat elevated at higher temperatures and humidities. It is interesting to note that the three highest levels corresponded to the three dwellings with the lowest air permeability levels (H7. H18. All results were less than the WHO guideline level for effects on health and comfort of 100 μg.9 Formaldehyde This section provides a summary of the results of the formaldehyde (HCHO) levels measured in the living room and master bedroom. It would be expected that there would be some variation in these three days such that there would be 30-minute periods that would be higher. Figure 18 provides the distribution of the highest formaldehyde levels monitored in each home. then the rate of emission will be quite steady. it would suggest that at most three dwellings would have exceeded the WHO criterion. Assuming the sources are predominantly formaldehyde based resins.10 in the Appendix. in wood based products. H20).g. Formaldehyde levels monitored during the study were averaged over three days. The highest level recorded during the monitoring period was 63 μg.Chapter 6 Results | 39 6.m-3 (WHO 2000) averaged over 30 minutes. Assuming that this would result in maximum 30-minute levels double that of the three-day readings. Pollutant concentrations are more likely to vary due to variations in ventilation rate – there are likely to be some periods of significantly lower ventilation rate due to falls in external driving forces. Figure 18 Highest formaldehyde concentration in each dwelling . e. For results of individual dwellings see Table A.m-3 in a master bedroom.

for example through respiration. the exceedance occurred in the bathroom and not the kitchen. e.g. These occurred in the living room for H16 and in the bedroom for H11 and H20. This is to both minimise the pollutant levels in those rooms and to minimise the spread of pollutants to the rest of the building. washing-up. H16. In each case. Excessive levels were recorded in the wet rooms of three homes (H15. However. the results may be at least partially explained by the low air exchange rates in these homes. . The results from this study suggest that the intermittent extract flow rates may be insufficient to control moisture levels generated in the wet rooms.1 Are intermittent extract flow rates sufficient? The purpose of intermittent extract ventilation is to provide extract ventilation in rooms where there are significant indoor pollutant sources. the monthly RH levels were also high in the habitable rooms of three homes.40 | Ventilation and Indoor Air Quality in Part F 2006 Homes Chapter 7 Discussion 7. Extract ventilation is located in the kitchen to remove combustion pollutants from cooking and other moisture generating activities. Insufficient extract ventilation is likely to be one factor allowing dispersal to the rest of the home.24 ach) and would tend to generally increase the relative humidity level throughout the home. The air exchange rate in all three of these homes was low (0. low exchange rates are expected to have a significant influence on these results.27 ach). In two of the homes. H20) against the monthly RH criterion. Extract ventilation is also located in bathrooms to remove moisture generated in the air from bathing and showering. However.20 – 0. Again. RH levels in habitable rooms are controlled by a combination of extract ventilation to minimise dispersal from wet rooms to the rest of the dwelling and background ventilation to remove moisture produced elsewhere in the dwellings. in all three homes it is noted that the air exchange rate is low (0. Based on the same assumptions. the bathroom extract rates actually met CLG recommended levels (H15 and H20). This analysis is based on the reasonable assumption that the actual monthly levels in these homes were similar to the recorded weekly levels.20 – 0.

It is worth noting that of the four dwellings. as in many cases intermittent extract ventilation and/or trickle ventilator areas did not meet ADF 2006 guidance.Chapter 7 Discussion | 41 Further analysis considered a reasonably worst case external moisture scenario.h-1. H15. all apart from one of the dwellings had at least one bedroom unoccupied during the monitoring period. Overall the results from this study suggest that the intermittent extract flow rates are sufficient and should not be changed. the moisture levels would have been reduced in these rooms but spread and diluted throughout the rest of the dwelling. These may not have occurred in practice because of user behaviour (e. However.) during moisture production.7. (ADF assumes two people sleep in the master bedroom and one person sleeps in each other bedroom. although the sample size is very small to make any conclusive judgement based on this. A further dwelling did not use their cooker hood as it was too noisy. Focusing particularly on the wet rooms here (background ventilation is discussed further on).m-2. as discussed in Section 6. lower moisture production during cooking than predicted) or relatively low occupancy. in practice. but. either the intermittent extract could have been used for a longer period or windows could have been opened for short periods. In these cases. it is important that cooker hoods and extract fans are properly installed. it does suggest that if the cooker hood had been used. nitrogen dioxide levels would have been controlled. levels exceeded those recommended by ADF. One dwelling (H8) exceeded the 75% RH ADF 2010 criterion for weekly average indoor RH levels and three other dwellings approached this level (H2. Whilst it is important to recognise this problem. cases moisture generation rates. The intermittent extract flow rates were not sufficient to control nitrogen dioxide levels. It is also worth noting that these highest levels always occurred in the bathroom. However. . It is perhaps surprising that the RH levels were not higher more generally. three of the dwellings had extract flow rates approximately half that recommended by ADF and if the fans had been properly installed this should have addressed this problem.g. there are two plausible explanations. In four dwellings. This includes the more airtight dwellings in the sample. Hence. H20). three had air permeability below 5 m3.

can never be 100% effective (e. if the kitchen door is open during cooking.g. is to remove non-localised pollutants spread around the dwelling as well as remove pollutants that ‘escape’ extract ventilation which. The pollutants’ levels are colour coded: . There may be problems from formaldehyde in airtight homes – this is a relatively small study and care should be taken in extrapolating the results to the whole new building stock.1 Are trickle ventilator areas sufficient? Introduction The purpose of background ventilation. The table has been ordered from lowest to highest air permeability. Table 3 compares air permeability against pollutant levels. We shall next review whether TVOC levels are high because of inadequate installation and inspection or because the ventilation areas recommended in ADF 2006 are insufficient.2 Pollutant levels versus airtightness We first review the evidence for a correlation between the air permeability and indoor pollutant levels. as there was a greater prevalence of TVOC levels exceeding the IAQ criteria.2. The table focuses on pollutant levels in habitable rooms (ie extract fans should address high pollutant levels produced intermittently in wet rooms). It is worthwhile reviewing this relationship given the purpose of the study. this should suffice for RH levels as well. of which trickle ventilators are a means to achieving it. some cooking pollutants will be extracted by the fan and others spread to the rest of the dwelling through the open door). The formaldehyde levels from sources around the home were sufficiently low. In doing so. the RH and TVOC levels exceeded consultation ADF 2010 pollutant criteria. This analysis focuses on TVOC levels. in practice. by adequately controlling RH and TVOCs. we have not undertaken detailed statistical analysis. it may significantly control other pollutants as well. However.2. Due to the relatively small sample size. Of all of the pollutants monitored in the habitable rooms. rather than RH levels. It is assumed that if TVOC levels were adequately controlled. some approximations will need to be made as it is difficult to know exactly the impact of inadequate installation and inspection. 7.42 | Ventilation and Indoor Air Quality in Part F 2006 Homes 7.2 7. We shall first review the correlation of pollutants with airtightness.

34 0.5 19 24.89 4.2 0. As highlighted in Section 6.70 0.8 Worse weekly average % RH 74.49 0.8 58.45 3.0 61.1 22.9 μg.13 5. Furthermore.43 0.5 67.23 0.96 3.61 11.Chapter 7 Discussion | 43 The table suggests.84 9.22 5.27 0.25 5.3 n/a n/a 39.55 0. pollutant emission rates will vary between dwellings.51 10.m-3 535 n/a n/a 976 1213 224 572 311 n/a 111 204 191 330 464 n/a n/a 156 n/a n/a 53 231 320 μg. It is not surprising that there is not a clear correlation between airtightness and indoor pollutant levels.32 6.m-2 @ 50 Pa @ 50 Pa H20 Flat H15 Flat H11 Flat H7 Terrace-End H18 Flat H1 Detached H19 Flat H8 Terrace-Mid H12 Semi-Det H5 Terrace-End H6 Terrace-Mid H10 Terrace-Mid H9 Terrace-End H2 Detached H14 Semi-Det H16 Semi-Det H3 Detached H13 Detached H17 Detached P1 P2 Semi-Det Semi-Det 2.47 0.02 6.7 62.6 4.77 9.12 4.4 61.8 52.52 4.6 56.26 0.67 11.m-2 m3.28 8.37 7. However.97 0.8 15.5 63.45 0.37 3.8 42.89 4.44 0.7 50.42 0.3 13. the installed capacity of the ventilation systems shows significant variation.45 7.49 6.9 Max HCHO Max m3.38 0. Relatively high levels are observed in the five most airtight dwellings.3 43.37 0. there is a correlation between air permeability and indoor pollutant levels.2 65.09 7.35 7. it is thought that the flats may be significantly under-ventilated due to them having effectively a single façade but the ventilation system being designed for a multi-sided façade.32 4. as highlighted earlier.53 13.82 8.5 n/a 58.6 69.9 46.18 8.5 46.37 5.23 0.5 H4 Detached .7 n/a n/a 55.1 24. Table 4 still suggests that even with the flats removed.3 19.6 16.0 53.2 75.71 Ventilation RH (habitable rooms) TVOCs Weekly average % RH 71. it is less clear.0 53.h-1.2 28.8 n/a 57.11 11.4 49. given the limitations of the sample size. the anticipated correlation between air permeability and pollutant levels.99 0.5 66. However.38 0.9 73.4 29 11.h-1.81 7.62 0.9 13. Table 3 Comparison of air permeability and pollutant level Property details ID House type Air permeability Test Method A Test Method C Ach 0.9 58.25 9.3 13.1 56.7 31 31 32 46 19.1 4.3 66.2 65. For example. Four of the five more airtight dwellings are flats.02 7.m-3 50 55 37 63 35 30.22 7.98 5.1 41.74 14.0 62.58 8.8 56.24 0.4.67 8. there may be a confounding factor here.6 73.92 4.2 54.96 4.7 66.5 63.

11 11.6 6. compared to other parameters.89 4.25 6.7 66.h-1.m-2 @ 50 Pa @ 50 Pa H7 H1 H8 H12 H5 H6 H10 H9 H2 H14 H16 H3 H13 H17 P1 P2 H4 Terrace-End Detached Terrace-Mid Semi-Det Terrace-End Terrace-Mid Terrace-Mid Terrace-End Detached Semi-Det Semi-Det Detached Detached Detached Semi-Det Semi-Det Detached 3.02 7.1 24.51 10.47 0.8 50.89 4.67 11. The Table highlights the following: 3 .84 9.37 5.h-1.2 0.44 0.45 7.5 66.8 n/a 57.49 6.28 8.9 73.8 Worse weekly average % RH 66. Perhaps of more concern is that the other three were leakier and.3 Review of TVOC levels First. there is a suggestion that the high TVOC levels occur in more airtight properties.55 0.1 41.62 0.13 5.2 22.m-2 @ 50 Pa As before.18 8.37 0.98 5.34 0.12 4.3 13.7 31 31 32 46 19.4 75.9 μg.m-3 μg. it is informative to review the TVOC levels in more detail.71 Ventilation RH (habitable rooms) TVOCs HCHO Weekly average % RH 53.8 15.3 13.4 58. .6 16.81 7.0 53.5 62. if they had been more airtight.2 54.38 0.67 8.26 0.4 29 11.h-1.38 0.8 56.h-1.22 7.53 13. The levels are shown in Table 5.43 0.5 63.2.49 0.1 56.58 8.9 Max Max m3.37 7. the ventilation would be expected to be less and consequently the TVOC levels would have been higher.5 67.6 56.61 11.8 42.74 14.82 8.m-3 976 224 311 n/a 111 204 191 330 464 n/a n/a 156 n/a n/a 53 231 320 63 30.70 0.0 62. from this limited sample size.3 43.5 n/a 58. The data have been sorted by TVOC level.2 65.97 0.8 52.22 7. The five of the eight properties with high TVOC levels had air permeabilities below 5 m3.9 13.5 19 24.7 n/a n/a 55.5 46.44 | Ventilation and Indoor Air Quality in Part F 2006 Homes Table 4 Comparison of air permeability and pollutant level (without flats) Property details ID House type Air permeability Test Method A Test Method C Ach 0.45 0.3 n/a n/a 39.25 9.3 19.96 4.m-2 m3.99 0.m-2.9 46.35 7.5 7.77 9.

toluene.9 8.1 6.3 5.6 5. not all airtight properties are expected to have high pollutant levels. how would the TVOC levels vary? .3 9.62 0.7 7.7 0.2.m-3.2 13. If sufficient trickle ventilation was installed.3 5.5 3.m-2 @ 50 Pa 9. formaldehyde.47 0.38 0. having acceptable VOC levels.37 0.71 0.9 11.5 5. e.m-3 % 0 0 0 0 0 0 0 4 7 10 55 78 91 225 304 Method A m3.4 6.2 14. with an air permeability below 5 m3. It is also noted that three of the four highest level homes are flats which is explored further below.55 0.m-3 P1 H5 H3 H10 H6 H1 P2 H8 H4 H9 H2 H20 H19 H7 H18 Semi-Det Terrace-End Detached Terrace-Mid Terrace-Mid Detached Semi-Det Terrace-Mid Detached Terrace-End Detached Flat Flat Terrace-End Flat 53 111 156 191 204 224 231 311 320 330 464 535 572 976 1213 Ach 0.4 Impact of installing inadequate trickle ventilation This next step is to estimate the impact on the TVOC levels from the fact that insufficient trickle ventilation was installed in some homes. If TVOCs are less than the guideline level of 300 μg.23 0.8 11. then typically individual organic compounds (e.26 0.5 7.1 10.42 7. This is a possible explanation for the three dwellings.34 0.m-2.0 4.6 4.8 4.5 6.44 0.3 7.h-1.h-1.g.8 7. it is important to note that TVOCs were chosen in ADF to represent organic compounds as it was to be the most sensitive marker for organic compounds.Chapter 7 Discussion | 45 As referred in the previous sub-section.5 3.0 Ventilation μg.m-2 @ 50 Pa 13.h-1.0 8.1 4.9 4.3 5.49 0. Finally.g.38 0.4 5.24 0. benzene) should also be below health-based guidelines for these specific compounds.1 Method C m3. Table 5 Review of TVOC levels House House type ID TVOC Percentage above 300 μg. it depends on the presence of sources.

H4 and H8 are now less than or equal to 300 μg. Energy and Buildings. Ventilationcorrected = k. rate at 50 Pa. TVOC levels are increased where previously there was an excessive trickle ventilation rate compared to ADF 2006. This assumption is the basis of the 1/20th rule of thumb (see Sherman. Overall. Taking these two assumptions. could be expected on a different week’s monitoring period. In this analysis. doubling any reduction noted here would still not bring the high TVOC results significantly below 300 μg. It is the standard formula for pollutant concentration in equilibrium. . M H (1987) Estimate of infiltration from leakage and climate indicators. For comparison. the TVOC levels are reduced where there was previously under-installed trickle ventilation. it is not significantly reducing those homes with the highest TVOC levels.m-3. but this is still a cause for concern as they are still approaching the guideline level (and. if the correct amount of trickle ventilation was included as follows. 81–86. This is reasonable. We can then estimate the increase in ventilation rate. the TVOC pollutant concentration (ie ventilation rate = 1/TVOC level). 10. while this does have an impact. H9 and H10 due to a lack of equivalent area information on the trickle ventilator to determine the degree of any under or over installation. Hence: Ventilation = k. ACH50 where k = the proportionality constant. pp. respectively. Even if the above assumptions are underestimates. We already know the air pressure test results with and without the trickle ventilation open. 2 Sherman. 19872). the results are shown in Table 6. P2.m-3. (( trickle vent (ADF) trickle vent (installed) x [ACHC – ACHB]) + ACHB ) where ACHB and ACHC are the air leakage rate results at 50 Pa for Methods B and C. to exceed it). if at all.46 | Ventilation and Indoor Air Quality in Part F 2006 Homes In doing this analysis. we make a number of assumptions. Note that the analysis was not possible for dwellings P1. It assumes that the outdoor concentration has a small impact on the internal levels which is suggested by the results in this and other studies.

m-3.2.h-1.4 4. the impact of a more airtight property was determined applying the same techniques as in Section 7.m-2.m-2 @ 50 Pa P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H18 H19 H20 Semi-Det Semi-Det Detached Detached Detached Detached Terrace-End Terrace-Mid Terrace-End Terrace-Mid Terrace-End Terrace-Mid Flat Flat Flat 9. The highest formaldehyde levels were for the more airtight dwellings and hence only a small increase.6 4.1 4.4.9 5.5 Building more airtight dwellings It is also interesting to see the potential impact if all of the dwellings were built at least to 4 m3. if at all. If all dwellings had an air permeability built to 4 m3. is observed by assuming an air .2. This is shown in Table 7 below.m-3.5 3.6 5.Chapter 7 Discussion | 47 Table 6 Analysis of TVOC results due to incorrectly installed trickle ventilation capacity House ID House type Method A m3. analysis showed that this does not tend to take the levels above 100 μg.h-1.0 5.3 11.m-2 or better.5 5.m-3 Change % 7. Hence for those properties leakier than this. It should also be noted that this would also impact on the other pollutant levels.5 10.0 TVOC (original) μg.5 8.9 4.m-3 53 231 224 464 156 320 111 204 976 311 330 191 1213 572 535 1126 530 472 -7 -7 -12 185 488 172 300 109 198 890 296 -17 +5 +10 -6 -2 -3 -9 -5 TVOC (new) μg. nine homes would now exceed 300 μg.1 3.1 6. The results suggest very significant increases in TVOC levels.h-1.

9 4. they were essentially single sided. The air permeability has a greater impact on internal sources – an improved air permeability stops moisture from escaping the dwelling and.5 10. therefore. However. percentage increase would be less than that for TVOCs and formaldehyde.1 3.6 Inadequate ADF guidance in flats There is a suggestion that a contributing factor in flats is inadequate guidance in ADF. Table 7 Impact of improving airtightness on TVOC levels House ID House type Method A m3. Flats 18 to 20 had two external façades. To a first approximation.m-2.h-1.48 | Ventilation and Indoor Air Quality in Part F 2006 Homes permeability of 4 m3. the levels were fairly low to start with and.0 5.m-2 @ 50 Pa P1 P2 H1 H2 H3 H4 H5 H6 H7* H8 H9 H10 H18 H19 H201 * TVOC (original) μg.5 3. increasing the indoor levels.h-1.m-3 53 231 224 464 156 320 111 204 976 311 330 191 1213 572 535 TVOC (new) μg. Indoor moisture levels arise from both indoor and external sources.1 4.5 8. the results were still relatively low.h-1.m-2. in that one of the façades was short in length and thus little cross ventilation would occur .1 6. For those leakier dwellings. even with increases.3 11.5 5.9 5.6 4.0 These dwellings already had an air permeability better than 4 m3.2.6 5. hence the trickle ventilation design reflected this. so no analysis was undertaken 7.m-3 91 445 228 620 248 653 127 241 355 403 231 1237 637 Change % +71 +93 +2 +33 +59 +104 +14 +18 +14 +22 +21 +2 +11 Semi-Det Semi-Det Detached Detached Detached Detached Terrace-End Terrace-Mid Terrace-End Terrace-Mid Terrace-End Terrace-Mid Flat Flat Flat 9. the external sources provide a background level whatever the air permeability.4 4.

Chapter 7 Discussion | 49 in practice. Door undercuts were not measured in the pilot homes and hence these results are not presented here. . 444 and 331 μg. the impact of door undercuts depends on whether internal doors are left open and whether there are any other routes for air to move around the dwelling (i.m-3. this would reduce the TVOC levels by a further 15–30%. internal leakage through gaps. It is difficult to assess the impact here. there will be different source emission rates in different homes.e.4.7 Impact of door undercuts This is to consider the impact of undercuts being less than recommended by ADF2006. respectively.2. 7. Assuming that better guidance would result in a doubling of the ventilation rate from purpose-provided ventilation. As shown in Table 8. This lack of a clear correlation can be explained by several factors. Again. cracks separating internal rooms). The results would be for H18–20: 893. Furthermore. It may have been better if the ventilation guidance for single-sided façades was followed or another ventilation system adopted. there appears no clear correlation between door undercuts and TVOC levels.2. according to similar calculation as in Section 7. Just over 50% of the door undercuts did not achieve 10 mm and only one dwelling had all of its door undercuts achieving this target.

2. it is likely to result in nine of the 15 homes with TVOC levels above 300 μg.h-1. there is the suggestion that additional ventilation is needed for more airtight dwellings for the following reasons. tightened to 4 m3. However.m-2 or tighter (ie those homes for which ADF 2006 does not provide guidance) need additional ventilation compared to that recommended in ADF 2006.m-3 recommended in ADF 2006.m-2 appears to significantly outweigh likely effects from under installation of trickle ventilation. Furthermore. 300 μg. better guidance is needed for single-sided ventilation. in the case of flats. overall. Furthermore.h-1.50 | Ventilation and Indoor Air Quality in Part F 2006 Homes Table 8 Impact of door undercuts House ID House type TVOC μg.8 Discussion From this study. have been significantly under-ventilated as they were designed for two- .m-3 H5 H3 H10 H6 H1 H8 H4 H9 H2 H20 H19 H7 H18 Terrace-End Detached Terrace-Mid Terrace-Mid Detached Terrace-Mid Detached Terrace-End Detached Flat Flat Terrace-End Flat 111 156 191 204 224 311 320 330 464 535 572 976 1213 Undercuts (all rooms) mm 9 16 9 4 11 7 12 7 11 7 9 5 9 Undercuts (habitable rooms only) mm 6 16 14 5 13 7 13 7 13 10 12 5 11 7. This is mainly due to the need to estimate for effects of inadequate installation and inspection.m-3. it is not possible to provide conclusive evidence that dwellings built to 3-4 m3.

However. It is thought that this resulted in. Two homes exceeded the maximum level of 10 m3.Chapter 7 Discussion | 51 sided ventilation but ventilation was effectively only single-sided.m-2 required by ADL 2006 for dwellings sampled. It is not possible to know if these dwellings exceeded 10 m3.m-3. However. this still leaves five to six dwellings with levels above 300 μg.m-2 on completion. at least. Hence. in ADF. 7. the nitrogen dioxide levels exceeding WHO guidelines in at least three dwellings.4 Are dwellings Part L compliant? All dwellings tested are Part L compliant in that they have been given approval through the building control process. the study has demonstrated many cases where the recommendations given in ADF 2006 have not been met. Every dwelling had at least one fan or cooker hood that did not provide the recommended extract rate.h-1. occupants do not use their ventilation system to full capacity (trickle vents are not always open and fans are not always used).3 Are dwellings Part F compliant? All dwellings tested are Part F compliant in that they have been given approval through the building control process. it is unclear from the results the benefit that this would have. Finally.h-1. It is possible that the trickle ventilators. However.m-3. There is an additional confounding factor. may not be achieving this on installation taking into account their actual surroundings which could impede flow. it is important to note that in practice. The key differences from the ADF recommendations are as follows: ADF. This is something that should be considered in further work. 7. while producing the equivalent area in the laboratory. It is not clear whether better design would have reduced levels below 300 μg. the actual pollutant levels in dwellings will likely be higher than those recorded in this study. as they may not have been included in the .

h-1.m-2). undertake air pressure tests at lower pressure differences) may make it a useful technique. 7. There is relatively little difference between the two sets of results (mean of 0. further development of this approach (e. 7. is not able to determine the equivalent area with sufficient accuracy.5 Should we change air pressure testing for Part L compliance? These results question the need to move from BS EN 13829 Method A to Method B.5 m3. to ensure that there is some ventilation at their minimum setting. This study has not demonstrated the robustness of the approach.h-1. .m-2.52 | Ventilation and Indoor Air Quality in Part F 2006 Homes developer’s air permeability test sample. maximum of 0. at present.h-1.g. However.6 Should we undertake air pressure testing for Part F compliance? The study investigated whether it was possible to determine the total trickle ventilation equivalent area through testing in-situ. This may become more prevalent as homes become more airtight due to concern that occupants will close trickle ventilation in airtight homes because of inadequate information and result in poor indoor air quality and potential health effects. The advantages of making this change are as follows: of the dwelling. The approach. This needs to be balanced by the additional time and effort to seal and unseal all natural ventilation openings given the relatively small air permeability reduction. It is feasible that the dwellings achieved 10 m3. The proposal was to undertake air pressure testing with the trickle ventilators sealed and then opened and through analysis of the resultant data determine the equivalent area.2 m3.m-2 on completion and the air permeability increased post-completion due to drying out of the properties and/or work undertaken by the occupants. This will become more important as air permeability further improves and the relative contribution from non-sealed trickle ventilation increases.

. In the worst case. It is important to note that in developing ADF 2006. that recommended in ADF 2006.h-1. toluene. benzene. it would be expected that approximately four homes would have monthly relative humidity levels that would exceed the newly proposed monthly relative humidity guideline level. Furthermore. recommended in ADF 2006.g. only just over half (57%) of the recommended trickle ventilation was installed. A review of indoor air quality results suggested that if the TVOC guideline level of 300 μg. then individual chemical health-based guideline levels (e. the new daily and weekly relative humidity guideline levels proposed in the consultation version of ADF 2010. none of the fans located away from the cooker achieved the recommended flow rates in the kitchen.m-2.Chapter 8 Conclusions and recommendations | 53 Chapter 8 Conclusions and recommendations The key results from this study can be summarised as follows: ventilation and intermittent extract. that recommended in ADF 2006. 50% of main sample) achieved an air permeability of less than 5 m3. Ten of the 22 dwellings tested (46% of total sample including pilot homes.m-3 was met. sufficient trickle ventilation installed to meet ADF 2006. TVOCs were chosen to represent organic compounds. A key issue in these cases was insufficient installed flow where fans are located away from the cooker. formaldehyde) should also be met. making reasonable assumptions. However. recommended in ADF 2006.

While the flats had two façades. at least a significant minority of TVOC levels would be expected to exceed the recommended guideline level. particularly for the more airtight homes. the extract rates should be increased for more airtight dwellings. Hence. the TVOC criterion was selected as a sensitive marker for individual organic chemical compounds (ie if TVOC levels are below the criterion. The ventilation and indoor air quality results suggest that these homes were under-ventilated. but the long term effects on health due to exposure to VOCs are not well documented. that they are installed correctly to provide the capacity recommended in ADF 2006. which are associated with respiratory illnesses including asthma. There is no evidence from this study that with correct installation. VOCs are produced by building products and the activities of the occupants such as smoking. It may seem pedantic to change the design of a ventilation system just to better control one pollutant. the actual pollutant levels in dwellings will likely be higher than those recorded in this study. such as house dust mites. There are also other hazards to health that are affected by ventilation. previous research suggest that individual organic chemical compounds would also be below recognised indoor or outdoor health-based levels for these pollutants). system to full capacity (trickle vents are not always open and fans are not always used). Furthermore. installation issues for the ventilation system as a whole. Furthermore.54 | Ventilation and Indoor Air Quality in Part F 2006 Homes The key conclusions from these results are as follows: recommended natural ventilation provisions in ADF 2006 are sufficient for airtight homes. We do not currently attempt to control these under Part F because their breeding success is influenced by heating and hygiene practice as much as ventilation. and were ventilated as ‘multi-sided façades’ according to ADF 2006. and it seems prudent to err on the side of caution. the capacity of the ventilation system did not meet that recommended in ADF 2006. Improving ventilation would be a step in the right direction to limit this risk. The study is relatively small and the conclusions should only be treated as indicative. in all cases. . use of personal hygiene products and interior decorating. they were effectively single façades as the second façade was limited.

However. given to reducing source strength by product controls.h-1. This is based on the levels of TVOCs observed. to better determine indoor air quality levels achieved in airtight dwellings in which the ventilation system is installed correctly. .Chapter 8 Conclusions and recommendations | 55 The key recommendations from this study are as follows: inspected to provide the ventilation capacity as designed. and the implications for individual organic compounds. there are sufficient grounds to support the proposal to increase trickle ventilators in dwellings having an air permeability equal to or tighter than 4 m3.m-2. This could also have benefits in controlling other hazards to health. but this is difficult to quantify. findings of this study. this study.

10 6.90 A-B Difference m3.14 0.0 3.30 5.22 0.14 0.60 4.11 9.81 11.18 3.0 5.60 Air permeability Test Method B m3.h-1.48 3.61 4.54 4.30 13.02 3.0 5.89 5.3 2.53 14.1 4.m-2 @ 50 Pa 13.99 3.89 4.9 4.98 5.09 4.4 3.45 4.2 7.13 3.m-2 @ 50 Pa P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max Semi-Det Semi-Det Detached Detached Detached Detached Terrace-End Terrace-Mid Terrace-End Terrace-Mid Terrace-End Terrace-Mid Flat Semi-Det Detached Semi-Det Flat Semi-Det Detached Flat Flat Flat 9.10 0.81 7.98 6.1 3.19 Test Method C m3.32 14.47 0.3 .22 A-B Difference % 3.58 7.43 2.67 4.h-1.96 6.45 8.36 0.1 2.14 10.m-2 @ 50 Pa 9.67 11.73 3.37 3.25 7.7 1.49 5.43 0.5 2.h-1.11 0.32 8.82 4.32 7.37 7.04 0.85 4.35 13.4 5.4 3.18 8.96 8.28 11.4 4.90 7.0 2.29 0.80 5.80 11.16 -0.01 0.3 -1.73 5.14 0.15 0.53 0.51 10.92 8.22 8.2 4.88 4.30 4.7 2.77 11.59 5.37 0.37 6.96 11.23 6.23 7.48 0.6 4.09 0.10 0.56 | Ventilation and Indoor Air Quality in Part F 2006 Homes Appendix A Detailed results Table A.10 0.36 3.22 9.19 4.65 8.m-2 @ 50 Pa 0.10 4.1 Summary of air permeability test results Property details ID House type Test Method A m3.86 8.7 3.52 2.74 6.93 2.12 6.16 0.h-1.41 0.25 7.15 2.02 5.84 4.99 4.8 0.

92 0.000 35.500 41.16 1.62 1.94 0.000 45.600 64.000 35.80 0.470 21.73 0.900 75.000 30.000 35.470 33.2 Summary of trickle ventilation equivalent areas Property details ID House type Equivalent area Recorded mm2 P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max 1 2 ADF Recorded/ recommended recommended mm2 35.400 56.62 0.04 0.000 30.000 35.000 85.800 38.000 35.88 Semi-Det Semi-Det Detached Detached Detached Detached Terrace-End Terrace-Mid Terrace-End Terrace-Mid Terrace-End Terrace-Mid Flat Semi-Det Detached Semi-Det Flat Semi-Det Detached Flat Flat Flat N/A1 N/A1 53. Total geometric area of 57.000 40.750 25.95 0.470 Not measured in pilot study.73 0.500 25.62 1.63 1.000 50.000 95.73 1.770 40.73 0.62 0. .000 40.300 mm2 measured in both dwellings.Appendix A Detailed results | 57 Table A.000 35.000 30.000 35.000 60.470 41.300 42.000 45.000 30.500 27.500 40.000 0.75 1.000 35.000 N/A2 N/A2 25.700 25.10 0.79 0.40 0.000 35.92 0.000 40.

58 | Ventilation and Indoor Air Quality in Part F 2006 Homes Table A.3 Summary of trickle ventilation positions (recorded on initial survey) Property details ID P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Overall House type Semi-Det Semi-Det Detached Detached Detached Detached Terrace-End Terrace-Mid Terrace-End Terrace-Mid Terrace-End Terrace-Mid Flat Semi-Det Detached Semi-Det Flat Semi-Det Detached Flat Flat Flat 13 10 21 16 17 30 17 12 16 14 7 7 13 14 22 15 16 9 21 7 3 11 311 Number of installed trickle vents Trickle ventilator position Open 0 3 19 0 2 0 4 3 7 0 0 7 11 7 17 12 7 9 0 6 2 8 124 (40%) Closed 13 7 2 16 15 30 13 9 9 14 7 0 2 7 5 3 9 0 21 1 1 3 187 (60%) .

1 27.3 35.83 0.2 21.0 25.15 1.55 1.62 0.38 0.9 77.7 27.96 0.43 0.4 61.4 8.8 23.6 31.4 17.23 0.4 57.78 0.7 15.49 0.46 0.55 0.1 36.0 17.0 21.56 2.99 Air exchange rate PFT l/s 24.3 15.45 0.7 7.5 77.20 0.07 1.70 0.18 1.6 17.1 24.6 7.44 0.98 1.37 0.45 0.8 9.80 0.93 0.35 0.5 15.71 0.5 ADF Table 1.42 0.67 0.0 14.97 0.8 17.6 27.2 52.97 0.50 0.44 2.1b l/s 21.0 1.8 51.38 0.23 0.4 Summary of air exchange rate results Property details ID PFT ach P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max 0.8 21.44 0.50 0.06 1.1 19.01 0.6 26.47 0.4 26.5 21.0 23.4 35.1 18.07 Measured/ ADF recommended .27 0.2 11.2 17.47 1.26 0.20 0.34 0.0 27.08 1.02 1.1 36.5 19.2 31.99 0.0 52.5 11.24 0.Appendix A Detailed results | 59 Table A.

22 19.8 2.1 10.6 23.0 17.9 3.6 4.7 34.6 3. .8 6.5 32.0 8.2 32.0 9.3 11.0 28.5 Summary of mechanical extract rates Property details ID Kitchen hood/ fan above cooker l/s P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max 1 2 3 Extract rates Kitchen fan away from cooker l/s Bathroom En-suite l/s 3.6 34. Difficult to seal to undertake test.5 15.0 12.9 9.5 7.4 9.9 2.0 19.5 19.7 15. Tested without diffuser.0 3.0 WC l/s 6.3 16.0 1.0 34.4 28.4 3.0 14.4 1.0 5.1 16.7 2.0 31.7 5.8 17.2 7.1 18.3 3.0 34.4 l/s 4.3 Fan not working 25.12 10.4 17.0 32.5 11.0 Fan fitted but unable to achieve a sufficient seal for test.1 17.0 27.2 4.1 13.8 21.6 28.5 15.3 14.7 35.1 8.0 32.1 18.6 15.2 21.7 6.1 Could not test1 35.0 17.6 15.60 | Ventilation and Indoor Air Quality in Part F 2006 Homes Table A.03 25.5 21.2 5.5 7.1 10.3 3.

4 60.8 52.8 62.6 56.2 73.4 43.6 n/a 74.2 52.4 Relative humidity Bathroom % RH 48.0 49.0 46.7 63.6 50.1 71.9 50.4 39.1 52.0 60.1 48.5 48.7 54.1 57.8 61.8 58.3 48.3 53.3 Bedroom % RH 50.0 53.2 69.7 49.4 39.3 46.9 57.2 61.7 62.1 52.1 60.1 45.1 53.3 53.3 36.2 69.6 50.2 42.0 56.4 52.2 63.4 70.1 47.6 36.1 66.6 36.8 71.7 58.5 62.8 53.8 53.5 n/a 71.4 49.7 56.1 71.6 42.7 57.4 43.3 67.0 42.6 52.1 51.6 60.4 53.0 55.3 43.0 Kitchen % RH 45.6 55.2 78.0 55.1 39.6 58.8 54.8 53.8 52.4 37.5 66.6 Summary of average relative humidity levels over monitoring period Property details ID Outside % RH H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max 78.3 58.2 62.5 65.6 51.9 71.0 71.3 43.8 60.0 55.6 41.5 61.Appendix A Detailed results | 61 Table A.0 67.7 60.4 37.0 Living room % RH 46.8 54.6 48.1 46.0 41.6 68.9 46.9 46.6 70.2 73.1 66.6 67.7 67.5 44.0 .8 71.

5 60.4 59.2 55.3 58.2 60.1 60.3 73.2 65.5 56.9 49.6 56.7 50.6 54.7 59.9 49.2 38.3 58.8 60.3 63.3 52.2 Relative humidity Kitchen % RH 49.1 54.9 44.2 50.6 54.5 .7 66.2 45.5 38.0 43.9 51.4 60.9 46.7 73.8 58.8 73.4 64.3 60.2 64.7 66.6 70.2 66.4 70.3 49.2 66.9 55.0 46.8 48.8 55.4 55.9 64.3 50.8 82.0 68.8 52.6 58.5 54.9 45.4 46.6 47.5 53.7 73.7 Summary of maximum average daily relative humidity levels Property details ID H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max Bathroom % RH 49.2 59.0 44.8 51.8 61.1 47.3 60.9 63.5 41.3 49.3 63.2 60.3 44.1 51.3 58.2 64.3 Living room % RH 51.9 58.2 56.8 45.9 65.5 55.9 62.9 56.1 Bedroom % RH 52.4 41.7 55.8 45.5 60.5 52.62 | Ventilation and Indoor Air Quality in Part F 2006 Homes Table A.8 82.

3 18.6 15.1 2.6 Outside μg.1 25.9 19. 1 Gas cooking.9 16.4 21.8 35.9 11.m-3 31.9 21.5 18.4 21.8 19.5 21.2 47.1 17.2 56.8 Nitrogen dioxide results (averaged over one week) Property details ID P11 P21 H11 H21 H31 H41 H51 H61 H71 H8 H9 H10 H11 H121 H131 H141 H15 H161 H171 H18 H19 H20 Mean Min Max * Sample missing.9 56.9 39.0 44.2 12.3 House type Semi-Det Semi-Det Detached Detached Detached Detached Terrace-End Terrace-Mid Terrace-End Terrace-Mid Terrace-End Terrace-Mid Flat Semi-Det Detached Semi-Det Flat Semi-Det Detached Flat Flat Flat . Nitrogen dioxide levels Kitchen μg.2 31.3 23.9 24.5 40.Appendix A Detailed results | 63 Table A.1 * 31.9 21.1 2.4 22.3 33.0 17.9 22.1 31.6 26.6 55.8 9.5 11.m-3 33.9 26.6 21.6 7.2 19.3 23.6 7.3 13.4 18.4 9.4 40.

m-3 Outside μg.m-3 P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max 1147 1213 103 112 90 114 708 219 251 191 n/a n/a n/a n/a n/a n/a n/a 1147 572 56 264 168 285 53 231 224 464 156 320 111 204 976 311 330 171 n/a n/a n/a n/a n/a n/a n/a 1213 446 535 379 n/a n/a n/a n/a n/a n/a n/a 103 TVOC levels Master bedroom μg.64 | Ventilation and Indoor Air Quality in Part F 2006 Homes Table A.m-3 .9 Total volatile organic compounds (TVOCs) results (averaged over one week) Property details ID Living room μg.

0 19.4 1.8 0.0 19.2 19.5 0.8 6.6 6.0 14.6 2.3 2.0 11.0 16.8 35.2 11.3 2.0 24.3 24.0 18.m-3 0.0 1.8 1.0 14.0 27.9 21.Appendix A Detailed results | 65 Table A.0 36.0 22.0 11.0 24.9 24.8 3.0 Formaldehyde levels Master bedroom μg.1 32.0 24.4 2.5 16.9 21.6 1.6 13.7 .2 63.0 46.9 12.3 0.0 63.2 55.5 19.0 28.3 31.2 31.4 37.0 54.5 15.2 25.6 1.8 2.m-3 15.1 0.0 30.0 1.0 28.7 1.4 29.7 37.0 Outside μg.m-3 P1 P2 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 Mean Min Max 10.1 50.4 54.10 Formaldehyde results (averaged over three days) Property details ID Living room μg.2 10.1 2.0 1.9 6.8 31.9 10.0 2.0 46.0 25.9 1.

ISBN 978-1409823759 ISBN: 978 1 4098 2375 9 Price £20 9 781409 823759 .