ArticlePDF Available

Building Airtightness: Research and Practice


Abstract and Figures

This report summarizes the state of the art on building air tightness by reviewing the current and recent literature on both research and practice. The focus of this report is on techniques to measure the tightness of the building envelope and on what has been learned by doing so. This report reviews over 100 of the most important publications relating to the topic. The report covered the fundamentals of air leakage including the hydrodynamics of leaks, which has led to all of the measurement techniques currently in use. The measurement techniques reviewed focus on the fan pressurization technique and its derivates, but the report covers novel techniques as well. Air tightness metrics allow data to be shared and compared and the basic air tightness metrics are reviewed and discussed as well as a brief discussion on norms and normalization. The bulk of the report discusses data which has been taken over the last twenty years and what it can tell us about buildings of different types, locations and properties.
Content may be subject to copyright.
Building Airtightness: Research and Practice
M.H. Sherman (
Rengie Chan (
Lawrence Berkeley National Laboratory
Berkeley CA 94720
This report summarizes the state of the art on building air tightness by reviewing the current
and recent literature on both research and practice. The focus of this report is on techniques
to measure the tightness of the building envelope and on what has been learned by doing so.
This report reviews over 100 of the most important publications relating to the topic. The
report covered the fundamentals of air leakage including the hydrodynamics of leaks, which
has led to all of the measurement techniques currently in use. The measurement techniques
reviewed focus on the fan pressurization technique and its derivates, but the report covers
novel techniques as well. Air tightness metrics allow data to be shared and compared and
the basic air tightness metrics are reviewed and discussed as well as a brief discussion on
norms and normalization. The bulk of the report discusses data which has been taken over
the last twenty years and what it can tell us about buildings of different types, locations and
KEYWORDS: Air leakage; airtightness; fan pressurization; leakage area
This work was supported by the Assistant Secretary for Energy Efficiency and
Renewable Energy, Building Technology Program of the U.S. Department of Energy, under
Contract no. DE-AC03-76SF00098
ABSTRACT.................................................................................................................................... 1
TABLE OF CONTENTS............................................................................................................... 2
INTRODUCTION.......................................................................................................................... 3
S................................................................................................. 4
THE HYDRODYNAMICS OF LEAKS....................................................................................... 5
FAN PRESSURIZATION MEASUREMENT TECHNIQUES................................................. 6
................................................................................. 7
................................................................................................ 8
AIR TIGHTNESS METRICS....................................................................................................... 8
......................................................................................... 8
................................................................................................ 9
Norms and normalization ...................................................................................................... 10
AIR TIGHTNESS DATA ............................................................................................................ 11
............................................................................................................ 11
AIVC NUMERICAL DATABASE........................................................................................... 11
TRENDS BY BUILDING CHARACTERISTICS.................................................................... 14
ENERGY-EFFICIENCY DWELLINGS................................................................................. 15
KEY LEAKAGE PATHWAYS ................................................................................................ 16
IMPLICATIONS OF AIR TIGHTNESS MEASUREMENTS.................................................. 20
DWELLINGS .............................................................................................. 20
LOWER-RISE BUILDING MEASUREMENTS ..................................................................... 20
KEY LEAKAGE PATHWAYS ................................................................................................ 24
TRENDS BY BUILDING CHARACTERISTICS.................................................................... 24
BUILDINGS ......................................................................................... 25
TRENDS BY BUILDING CHARACTERISTICS.................................................................... 28
KEY LEAKAGE PATHWAYS ................................................................................................ 29
DYNAMIC AIR FLOW............................................................................................................... 30
.................................................................................................................. 31
SUMMARY .................................................................................................................................. 32
NOMENCLATURE..................................................................................................................... 34
ACKNOWLEDGMENTS............................................................................................................ 34
REFERENCES............................................................................................................................. 34
2 OF 46
“Air Tightness” is the property of building envelopes most important to understanding
ventilation. It is quantified in a variety of ways all of which typically go under the label of
“air leakage”. In this report we will review the state of the art of air tightness research.
Before reviewing what is known about air tightness, we will summarize the key roles air
tightness play in understanding ventilation.
Air tightness is important from a variety of perspectives, but most of them relate to the
fact that air tightness is the fundamental building property that impacts infiltration. There
are a variety of definitions of infiltration, but fundamentally infiltration is the movement of
air through leaks, cracks, or other adventitious openings in the building envelope.
The modeling of infiltration (and thus ventilation) is a separate topic, but almost all
infiltration models require a measure of air tightness as a starting point. While the magnitude
of infiltration depends on the pressures across the building envelope, the air tightness does
not, making air tightness a quantity worth knowing in its own right for such reasons as stock
characterization, modeling assumptions or construction quality.
Infiltration, and therefore air tightness, is important because it impacts building energy
use, and the transport of contaminants between indoor air and outdoor air (i.e. ventilation).
From an energy standpoint alone it is almost always desirable to increase air tightness, but if
infiltration is providing useful dilution of indoor contaminants, indoor air quality may suffer.
In many countries infiltration is the dominant source of outdoor air. Providing appropriate
IAQ at minimal energy costs is a complex optimization process that includes, but may not
be dominated by air tightness concerns. A high degree of air tightness will provide
insufficient air through infiltration and thus necessitates a designed ventilation system.
In buildings with designed ventilation systems, especially those with heat recovery, air
tightness may be a determining factor in the performance of that system. For example
unbalanced ventilations systems such as exhaust fans require that make-up air come through
building leaks. Overly leaky or overly tight buildings could reduce the effectiveness of such
When poor air tightness allows air to be drawn in from contaminated areas, indoor air
quality can be reduced even though total ventilation may be increased. These contaminated
areas could be attics, crawlspaces or even the outdoors. Sometimes the building envelope
itself may be a source of contamination because of mold or toxic materials. .
Moisture is a special class of contaminant because it commonly exists in both liquid and
vapor form and is a limiting factor in the growth of molds and fungus. Poor air tightness
that allows damp air to come in contact with cool surfaces is quite likely to lead to the
growth of microbiologicals. In cold climates poor air tightness can lead to the formation of
ice in and on exterior envelope components.
3 OF 46
Often the most noticeable impact of poor air tightness is draught and noise. Tight
buildings provide increased comfort levels to the occupants, which in turn can have impacts
on energy use and acceptability of the indoor environment.
From a measurement standpoint, air tightness means measuring the flow through the
building envelope as a function of the pressure across the building envelope. This
relationship often fits a power law, which is the most common way of expressing the data.
The power law relationship has the form
Blower Door” is the popular name for a device that is
capable of pressurizing or depressurizing a building and
measuring the resultant air flow and pressure. The name
comes from the fact that in the common utilization of the
technology there is a fan (i.e. blower) mounted in a door; the
generic term is “Fan Pressurization”. Blower-Door technology
was first used in Sweden around 1977 as a window-mounted
fan (as reported by Kronvall, 1980) and to test the tightness of
building envelopes (Blomsterberg, 1977). That same
technology was being pursued by Caffey (1979) in Texas (again
as a window unit) and by Harrje, Blomsterberg and Persily
(1979) at Princeton University (in the form of a Blower Door)
to help find and fix the leaks.
During this period the diagnostic potentials of Blower
Doors began to become apparent. Blower Doors helped
Harrje, Dutt and Beya (1979) to uncover hidden bypasses that
accounted for a much greater percentage of building leakage
than did the presumed culprits of window, door, and electrical
outlet leakage. The use of Blower Doors as part of retrofitting
and weatherization became known as House Doctoring both
by Harrje and Dutt (1981) and the east coast and Diamond et
al. (1982) on the west coast. This in turn led Harrje (1981) to
the creation of instrumented audits to computerized
While it was well understood that Blower Doors could be
used to measure air tightness, the use of Blower-Door data
could not be generally used to estimate real-time air flows
under natural conditions or to estimate the behavior of
complex ventilation systems. When compared with tracer-gas
measurements, early modeling work by Caffey (1979) was
found wanting. There was a rule of thumb, which Sherman
(1988) attributes to Kronvall and Persily that seemed to relate
Blower-Door data to seasonal air change data in spite of its
simplicity. Modeling of infiltration, however, is discussed
where C [m
] is the flow
coefficient and n is the pressure
exponent. The pressure exponent is
normally found to be in the vicinity of
0.65 but has the limiting values of 0.5
and 1 from simple physical
considerations. Because of the non-
linear nature of this expression there
are some interesting challenges in
understanding any measured data;
these issues will be addressed in
subsequent sections.
In her general study of air flow
measurement, McWilliams (2002)
reviews of the techniques for
measuring air tightness. The vast
majority of techniques fall into the
category of “fan pressurization” in
which a fan (or blower) is used to
create a steady state pressure
difference across the envelope. The
flow through the fan is measured at a
variety of pressures. The most
common incarnation of fan
pressurization technique for dwellings
and small buildings is known as a
blower door. Although other
methods for measuring air tightness
have been examined we shall concern
ourselves with principally with fan
pressurization techniques.
4 OF 46
Before discussing measurement techniques in any more detail, it is important to
understand the physical properties of the thing we are measuring, namely the leaks
Although the power-law has been found to be a reasonably good empirical description
of the flow vs. pressure relationship, it does not simply correspond to any physical
paradigm. There are physical paradigms that could be (and have been) applied to the
problem of air tightness:
If the leak is very short, frictional forces in the leak itself can be ignored and the
leak may be treated as a orifice in which the flow is proportional to the square
root of the pressure drop. The higher the flow rate (i.e. Reynolds number) the
longer the leak can be and still be treated as an orifice.
If the flow rate (Reynolds number) is low enough, the flow will be dominated
by laminar frictional losses and the flow will be linearly proportional to the
pressure drop.
Comparing to the power-law, the first case corresponds to an exponent of 0.5 while the
second case corresponds to an exponent of 1. The fact that measured data typically results
in an intermediate value indicates that neither of these two limits is a good explanation.
The Reynolds number of a typical leak is below that at which fully developed turbulent
flow is an issue, but the length of many such leaks is such that laminar friction is neither
negligible nor dominant. The problem becomes one of developing laminar flow in short
Sherman (1992) used the standard techniques for developing laminar flow to
characterize the problem of short circular pipes. In such a development the pressure drop is
the sum of that associated with the acceleration of the fluid and friction losses of the form:
P= Q+
This expression can be used to derive a quadratic relationship for flow as a function of
pressure, but the more interesting result is that it can be manipulated into a power-law
Where S is a dimensionless pressure:
5 OF 46
S= P
Where the exponent can be determined from S (or vice-versa):
If the leak were a single circular pipe, this derivation could, in principle, be used to
determine the diameter and length of the leak, but real envelopes are much more
complicated. Walker, Wilson and Sherman (1997) expanded this derivation to look at more
general crack geometries and the issue of series and parallel leaks.
The analysis assumes a smooth pipe. As shown by Kula and Sharples (1994) among
others, roughness can have a substantial impact and must be considered if the parameters of
this model are to be interpreted physically. The form of the model would only need be
changed if the roughness induced a transition to fully-developed turbulence in the leaks that
dominate the flow, but that has not been reported for real buildings.
The benefit of this analysis is not so much in providing an ability to infer the geometry
of leaks, but to confirm that a power-law formulation is a robust description on which to
base data analyses. It also tells us that the exponent is pressure dependent. This dependency
is low, so that over a narrow range of pressures the exponent can be assumed to be fixed. If
the pressure ranges over order of magnitude, however, one cannot assume it is a constant.
The fan pressurization technique has been around a long time and there are many
standard test methods that describe its use, such as ASTM (1999, 2002), CAN/CGSB (1986)
and ISO (1996). The basic technique involves measuring the steady-state flow through the
fan necessary to maintain a steady pressure across the building envelope.
The first level reporting of this data is generally the same. One reports the pressure and
volumetric flow at whatever measurement stations were chosen. If necessary, the raw
readings from the equipment may need to be corrected for zero offsets, temperature, altitude
etc. Such corrections are standard experimental practice, but will depend on the details of
the apparatus and experimental layout.
What separates the different test methods and protocols derived from them is the
analyses of that pressure-flow data. The simplest protocol and the one that is used most
often is simply to measure at a single pressure. The pressure chosen is conventionally 50 Pa;
so much of the published data quotes air flow at 50 Pa.
6 OF 46
As a metric air flow at 50 Pa has much to recommend it. 50 Pa is high enough to
overpower pressure noise and zero drifts caused by wind or stack effects. Thus it is
reasonably precise and therefore reproducible. The simplicity of a single-point measurement
and its reproducibility are why it is the most popular measurement.
Unfortunately, the flow at 50 Pa is not the quantity of interest if one is trying to
understand what envelope air flows are under natural driving pressures. The average
pressure across a leak in a building envelope is closer to 1 Pa than to 50 Pa. To have an
accurate estimate of air tightness is it necessary to determine it at normal pressures.
Furthermore, higher pressures can induce non-linear effects such as valving that would not
be relevant for normal pressures.
Depending on the metric chosen such reference pressures would be in the 1-4 Pa range,
but because these pressures are the size of the natural pressure variations, it is very difficult
to get a precise measurement of air flow. One must sacrifice precision to get accuracy or
must sacrifice accuracy to get precision.
In order to mitigate these errors, many test methods require that the flow be measured
over a range of pressures and then extrapolated to the reference pressure of interest using
the power law. Because of the non-linearities of the power-law and the biases that can be
associated with pressure measurements, care must be taken not to introduce unnecessary
errors into the data analysis. Modera and Wilson (1990) looked at the impact that wind
pressure variations have on the analysis of pressurization data and methods to mitigate them
using pressure averaging.
Sherman and Palmiter (1995) have examined the errors associated with analyzing fan
pressurization data including precision, bias and modelization errors. They examine the
overall uncertainty for a variety of analysis strategies and recommend optimal strategies for
selecting instrumentation and pressure stations.
The discussion above has focused on single-zone pressurization techniques. Although
such tests are vast majority of tests, in many circumstances the actual configuration is not
single-zone. Some of this is due to a true multizone nature, but some of this can be due to
the fact that there is no true air barrier between the “inside” and the “outside”.
Attached housing has leakage paths both to outside and to other dwelling units. Even
detached housing can have multizone properties when buffer spaces partially connect to the
living area and partially connect to outside. For detached housing the experimental problem
can often be solved by making a determination of what constitutes the air barrier and then
opening up doors and windows that are not part of the air barrier; thus reducing the
configuration to a single zone.
For apartments and other attached dwelling units, it is sometimes desirable to separately
know the leakage to the outside and the leakage to other adjacent units. Although not used
widely there are measurement approaches for determining these. Most methods such as that
7 OF 46
used by Levin (1991) in Sweden require access to adjacent units and often multiple blower
doors. Some researchers, e.g., Shaw (1980), have used a single blower-door and auxiliary
pressure measurements to infer component leakage.
Duct leakage measurement techniques are a spin-off from envelope air tightness
techniques. There are significant differences because of the fact that ducts operate under
externally applied pressure differences. When the air handling system is not operational,
duct leakage looks quite similar to envelope leakage and may represent a quarter of the total
envelope leakage.
The topic of air distribution leakage is too broad to be reviewed herein. Francisco (2001)
had reviewed five measurement techniques that have been under evaluation, but the field is
active and there have been developments since then. Carrie et al (1997) have looked at
some duct leakage issues an European context. A new standard test method in the U.S.
(ASTM 2004) makes use of the novel DeltaQ method for determining leakage.
Etheridge (1977) has been a proponent of the quadratic representation of flow, but most
researchers use the power law. In both cases, however, the representation is a two-
parameter model, with a recognition that these parameters may vary when the range of
applied pressure becomes large. Since Sherman (1992) showed that these representations
can be interchanged, we will only discuss the common, power-law representation.
Although there is general agreement that the power law is a good descriptor of air
tightness data, there is no real agreement on the best metrics to use in quoting air tightness
data. The best way to quote air tightness data will depend on what you plan to use it for.
Issues such as how many parameters to be used in quoting air tightness data and whether or
not air tightness data should be normalized by the size of the building are important when
deciding upon the optimal metric.
Whenever a two-parameter description of the air tightness is used, the second parameter
is always the power-law exponent, n. The exponent is critical for extrapolating measurements
from one pressure regime to another. When the actual measurements are made in the
pressure regime for which the data is desired—as often happens for 50 Pa metrics
extrapolation is not necessary and high accuracy determination of the exponent is
unnecessary. For such cases it is often sufficient to use the average exponent found from
large datasets, which has been found by Orme et al. (1994) to be approximately 0.65.
The exponent is also interesting from a research and/or diagnostic perspective because
it provides an indication of the relative size of the dominant leaks. If the leakage paths are
dominated by large, short leaks (e.g. orifices) one would expect the exponent to be closer to
8 OF 46
0.5; if the leakage is dominated by long-path leaks one would expect the exponent to be
closer to 1.
When making measurements before and after some retrofit or other sealing operation, it
is especially important to consider changes in the exponent. The exponent can be different
before and after such an operation. If an extrapolation is done without taking this into
account, the change in air tightness can significantly mis-estimated. Usually it is easier to seal
the large leaks, which tends to imply that a post-sealing measurement will tend to have a
higher exponent.
Whether found by extrapolation, interpolation or direct measurement, the principle
metric used to quantify air tightness is the air flow through the envelope at a specific
reference pressure. The most common reference pressures are 50 Pa and 4 Pa, but 1 Pa, 10
Pa, 25 Pa, and 75 Pa are used as well. The air flow is often denoted with the reference
pressure as a sub-script (e.g. Q
or Q
75 Pa was once suggested as a reference pressure because other envelope components
are sometimes tested at this pressure (e.g. windows (Henry and Patenaude (1998)). In
practice this pressure is too high to use both because some components may change under
that much pressure and because the pressurization equipment is often too small to achieve
that pressure directly. The air flow required to reach this pressure may itself be a problem
because of the flow required or in severe climates.
50 Pa, by contrast, is the most common pressure to measure the air flow. This has been
the traditional value since blower door techniques became popular. It is low enough for
standard blower doors to achieve in most houses and high enough to be reasonably
independent of weather influences. When single-point measurements are made, it is almost
always at 50 Pa.
25 Pa, is a standard reference pressure for measuring duct leakage (Cummings et al.,
1996). It is sometimes used as an envelope reference pressure for that reason. It is also
sometimes used as an alternate single-point pressure station when the equipment cannot
reach 50 Pa.
10 Pa is used as the reference pressure in the Canadian definition of equivalent leakage
area, but not normally directly as a flow rate.
4 Pa is similarly used as the reference pressure in the ASTM (E779-99) definition of
Effective Leakage Area (ELA) and in the ASHRAE Standards that reference it. ELA can be
defined as the area (of unity discharge coefficient that would have the same flow rate at the
specified reference pressure:
9 OF 46
where 4 Pa is chosen as the reference pressure as being representative of weather-
induced pressure
1 Pa is the lowest of the reference pressures used in the literature. At a pressure of unity
the power-law coefficient is equal to the flow rate. This form appears to make this metric be
independent of the power-law exponent, but because of the non-linearities and cross-
correlations associated with the measurement process, this is an illusion based on the system
of units used. Furthermore, extrapolation of the measured data, which is normally collected
at much higher pressures, is more uncertain than for any other reference pressure.
Flow rate at a specified pressure and leakage area at a specified pressure contain the
same information, just in different forms. Flow rate formulations are easier for those doing
the measurements because it relates more directly to their equipment. Leakage area
formulations are sometimes more intuitive for the occupant or owner because they can
imagine an amount of holes in their structure of a certain size.
The metrics above all refer to the total amount of leakage of the tested envelope. For
setting norms or standards, or for comparing one structure to another it is often desirable to
normalize this total by something that scales with the size of building. In that way buildings
of different sizes can be evaluated to the same norm.
There are three quantities commonly used to normalize the air leakage: building volume,
envelope area, and floor area. Each has advantages and disadvantages and each is useful for
evaluating different issues:
Building volume is particularly useful when normalizing air flows. When building
volume is used to normalize such data the result is normally expressed in air changes per
hour at the reference pressure; ACH
is probably the most common air tightness metric
reported. Many people find this metric convenient since infiltration and ventilation rates are
often quoted in air changes per hour.
Envelope area is particularly useful if one is looking to define the quality of the
envelope as a uniform “fabric”. Dividing (especially a leakage area) by the envelope area
makes the normalized quantity a kind of porosity. Although this normalization can
sometimes be the hardest to use, it can be particularly useful in attached buildings were some
walls are exposed to the outdoors and some are not.
Floor area can often be the easiest to determine from a practical standpoint. Because
usable living space scales most closely to floor area, this normalization is sometimes viewed
as being more equitable. This normalization is used most often with ELA measurements
and can be converted to a different kind of dimensionless leakage, such as the normalized
leakage used by ASHRAE (2001).
10 OF 46
Air tightness data can be expensive to collect. The larger and more complex the
building, the more difficult and time-consuming it is to collect the data. Furthermore, air
tightness in large buildings was not thought to be as important a consideration as for
dwellings. Thus, the majority of existing data is for dwellings and more specifically for
single-family homes. We shall review those first and then move on to the other kinds of
A report by Orme et al. (1994) describes the AIVC air tightness numerical database.
Over 2,000 measurements on single and multi-family dwellings are summarized. These data
were collected from ten countries as listed in Table 1. Mean air flow rates at 50 Pa are shown
by country in the report but it should be emphasized that they only act as guidelines because
air tightness can vary a lot from building to building.
Expected values for air tightness have been developed for number of generic forms
of construction, namely: timber frame and block-and-brick for low-rises, concrete/curtain
wall for high-rises, concrete panel and metal panel for industrial buildings. For each of these
construction types, the effects to air tightness from a number of building characteristics are
tabulated. For example, the ‘basic leakage’ for a low-rise building with a timber frame is
suggested to be 3 ACH
. If no vapor barrier is present, the dwelling is expected to be leakier
and the air leakage value should be increased by 3 ACH
. On the other hand, if the dwelling
has gasket window/door frames, then 1 ACH
should be subtracted from the default value.
Apart from these generic air leakage guidelines, Orme et al. (1994) also summarized
1,758 flow exponent measurements from Canada, Netherlands, New Zealand, the UK and
the US. The distribution of flow exponent is roughly normal with a mean value of
approximately 0.66. The authors did not observe meaningful relationship between ACH
and the corresponding flow exponent.
Factors that affect air tightness include age of construction, building type (single-family
versus multi-family dwellings), severe climate, and construction materials. Many of the
findings are confirmed by recent studies, which are discussed in more detail below.
Air tightness measurements of single-family dwellings are by far the most abundant
among the different building types. Many studies measured air tightness as a starting point
and then make use of the findings to address problems such as ventilation, energy cost, and
indoor air quality. There are also some focuses in research on air tightness of energy-efficient
dwellings and techniques to achieve higher level of air tightness.
11 OF 46
Air tightness is known to vary
greatly among dwellings. This is not
only true in countries where the
climate is relatively mild, such as that
in the US (Sherman and Dickerhoff,
1998) and the UK (Stephen, 1998),
wide variation has also been observed
in more severe zones, such as in
Canada (Parent et al., 1996) and
Sweden (Kronvall and Boman, 1993).
A ten-fold difference between the
leakiest and tightest dwellings has
been observed in those studies where
the size of sample is relatively large.
The same variation in air tightness is
evident even among new dwellings
according to studies in Canada
(Hamlin and Gusdorf, 1997), Belgium
(Wouters et al., 1997), and the US
(Sherman and Matson, 2001).
It is almost impossible to do a good job of
analyzing measurement data without an
understanding of the uncertainties that go along with
the measurements. Standard texts describe
considerations of precision and accuracy as well as
error propagation and robustness; such information
will not be repeated here. Sherman and Palmiter
(1995) have used these techniques to develop specific
expressions for fan pressurization and to optimize
the measurement process.
Few of the references in this section, however,
report rigorous uncertainty analyses. In fact, some of
the relatively early publications have included
incorrect error analyses because they failed to
properly account for the fact that the non-linear
nature of the power law, makes parameter errors
highly correlated. When this error happens during
an extrapolation it greatly increases the apparent
error (e.g. in the ELA).
Most of the reported data is based on single-
point measurements and assumed exponents. Using
extant exponent data as a prior in a Bayesian analysis,
it is possible in principle to estimate the extrapolation
bias caused by the assumed exponents, but this kind
of analysis is very rare.
In looking at large datasets, one hopes that the
central limit theorem will apply and that all of the
biases and other uncertainties will be reflected in the
standard deviations of the data themselves.
12 OF 46
Table 1 List of data sources and sample sizes in AIVC database and more recent studies.
Country AIVC Database Recent Studies
Sources Size Sources Measurements
Canada CMHC 475 Gusdorf (2003)
Hamlin (1997)
Buchan (1996)
Parent (1996), Proskiw (1998)
Buchan (1992), Fugler (1994)
Scanada (2001)
Elmahdy (2003), Proskiw (1995)
Fugler (1999)
Petrone Architects (2000)
Air-Ins Inc. (1998, 1998b)
37,490 mostly S-F
2,263 S-F Dwellings
(incl. 63 R-2000 Houses)
11 Log Houses
47 S-F Dwellings
Basements & Crawlspaces
Attached Garages
Air Barriers
Building Materials & Joints
US LBNL 435 Sherman (2001)
Desjarlais (1998), Yuill (1998)
Kosny (1998), Petrie (2003)
Brennan (1990)
Louis (1995)
Wilcox (2001)
70,000 S-F Dwellings
Exterior envelopes
ICF Systems
Air Barriers
UK BRE 385 Stephen (1998)
Lowe (1997)
McGrath (1996)
96 S-F Dwellings
15 2–Storey Dwellings
Sweden SIB 144 Sikander (1998) 3 S-F Dwellings
France CETE de Lyon 66 Litvak (2000) 37 S-F Dwellings
Belgium BBRI 57 Bossaer (1998)
Pittomvils (1996)
200 S-F Dwellings &
Apartment Units
6 Low-Energy Houses
Germany Zeller (1993) 48 S-F Dwellings &
Apartment Units
NO 303
New Zealand BRANZ 83
NBI 40
Switzerland NEFF, EMPA,
Ingenieur und
S-F denotes single-family dwellings
13 OF 46
Among the largest database to date on air tightness of single-family dwellings is the
LBNL Residential Diagnostics Database which has over 73,000 measurements from across
the US. Data collection is an ongoing effort by the Energy Performance of Buildings Group
at LBNL. A recent report by Chan et al. (2003) summarizes the measurements in terms of
year of construction, size of dwelling, presence of heating ducts, and floor/basement
construction type. The database also contains measurements from two special groups of
houses, namely energy efficiency programs and weatherization program for low income
Among the building characteristics mentioned, year of construction and size of
dwellings are found to be the most influential factors related to air leakage. The distribution
of normalized leakage is roughly lognormal. Regression analyses show that the geometric
mean of normalized leakage can be predicted by year of construction and size of dwelling.
Using regression analysis, additional variables were tested to see if the inclusion of them
improves prediction. Neither the location of dwelling, the presence of heating ducts, and the
floor/basement construction type was found to be significant. The result is a simple model
that can predict the air leakage distribution for a housing stock in the US using only
distributions of year of construction and size of dwellings as inputs.
Many studies have observed similar trend by comparing the air leakage of dwellings built
from different periods of time. Analysis based on over 2,000 houses showed consistent
increase in air tightness across all regions of Canada (Hamlin and Gusdorf, 1997). Kronvall
and Boman (1993) concluded similarly from an analysis of 50 single-family houses in
Sweden. The authors observed over 2 folds reduction in the mean ACH
of houses built
before 1940 and those that were built in 1976-88.
In countries where the maximum allowable air leakage for new dwellings is written into
building codes, e.g., Sweden, the reason for air tightness improvement over time is obvious.
However, in milder-climate countries where there is no air tightness standard or code on
new dwellings, newer dwellings are not necessarily more air tight than older ones. Stephen
(1998) analyzed the air tightness measurements of 471 UK dwellings carried out by BRE and
found no apparent systemic differences. On the other hand, voluntarily changes in
construction practices in the US have resulted in tighter buildings. Analysis on earlier
version of the LBNL Residential Diagnostics Database by Sherman and Dickerhoff (1998)
showed a clear decrease in air leakage from the oldest constructions to those that were built
around 1980. After that, air leakage is fairly constant with year built.
Age of dwelling is a measure of deterioration from wear-and-tear which can induce air
leakage. This is different from using year of construction as the measure which captures the
possible influence from change of building practices on air tightness. Recent constructions,
however, appear to be fairly resistant to age-induced leakages. A study by Bossaer et al.
(1998) showed that among the 51 Belgian dwellings built between 1990 and 1995, there is no
meaningful relationship between duration of occupancy and air tightness. Similarly, Proskiw
14 OF 46
(1995b) measured the air tightness of 24 houses over periods of up to three years and
observed no significant degradation.
Influence of building geometry on air tightness has been studied by Bassett (1985)
from measurements on 80 single-family houses in New Zealand. The author showed that
envelope area normalized air flow rate at 50 Pa increases as the geometry of the envelope
gets more complex. Envelope complexity is defined as the joint length between wall, floor,
and ceiling, divided by the envelope area. Chan et al. (2003) also observed that floor area
normalized leakage is a function of dwelling size. While it is speculated that larger dwellings
tend to have better constructions and therefore tighter building envelopes, the explanation
can also be that larger dwellings have more favorable surface area to volume ratios and/or
less envelope complexity.
Dwellings in severe climate such as Sweden, Norway, and Canada are known to be more
air tight than those that are located in milder climate such as the US and the UK. The main
reasons for tighter construction are to conserve energy cost and maintain thermal comfort.
Within Canada, Hamlin and Gusdorf (1997) observed consistent regional difference in air
leakage of houses built from different period of time. For a qualitative sense of how air
tightness of dwellings from different countries compares, Orme et al. (1994) showed up to
two to three-fold differences in mean ACH
among the ten countries listed in the AIVC
numeric database. The data used to compute those mean values included both single-family
and multi-family dwellings and are not adjusted for other influential factors, such as year of
construction. The findings nonetheless support the general notion that dwellings in more
severe climate are more air tight.
Few energy-efficiency programs in the US have specific air leakage performance
requirement. As a result, it is not clear whether the air tightness of energy-efficiency program
houses is guarantee, even though common practices of these programs, such as caulking and
weather-stripping, are known to help reduce air leakage. Persily (1986) measured the air
tightness of 74 passive solar homes located throughout the US and found little difference in
air tightness when compared to other dwellings in the country. At that time the data on
conventional houses being compared to were quite limited and cannot be considered as
representative of the US. It is nonetheless a surprising finding as noted by the author
because the passive solar homes were designed to consume relatively low levels of energy for
space conditioning, and were therefore expected to be more air tight.
More recently, Sherman and Matson (2001) compared the air leakage of new energy-
efficient houses against other new conventional houses. They found that energy-efficient
houses are tighter built in general, but the key benefit is that these programs promote
consistency in construction practice. This is demonstrated by less variation in the air
tightness of houses built under energy-efficiency programs compared to the others. In
15 OF 46
Canada, Hamlin and Gusdorf (1997) found that energy efficient R-2000
houses are at least
twice as airtight as new conventional houses in most regions of the country. However, the
gap between the two is narrowing as builders and house buyers are now generally more
aware of the problems associated with excessive air leakage.
There are also examples where consistency in construction practice is not realized by the
energy-efficiency program. In another air tightness comparison between 47 energy-efficient
residential buildings in New York State and 50 nearby conventional houses as controls, the
two groups have similar standard deviations (Matson et al., 1994).
The air tightness of low energy houses is particular important when the dwellings are
equipped with heat recovery ventilation system in order to achieve energy-efficiency.
Pittomvils et al. (1996) studied the air tightness of 6 low energy houses in Belgium for this
reason and found that the values ranged from 3.8 to 4.9 ACH
. Despite that these values are
half of those from conventional Belgian dwellings (Bossaer et al., 1998), air leakage at these
levels still compromise the fractional reduction in ventilation related building load. In
Germany, Zeller and Werner (1993) measured the air tightness of 48 dwellings where some
of them are designed to be low energy. About 40% of the dwellings tested have ACH
greater than 3 at which the ventilation system cannot be run energy efficiently.
The types of leakage problems have much to do with the construction of the dwellings.
In a project that studied the effectiveness of various retrofitting strategies, Lowe et al. (1997)
found that one of the most important factors is the method used to construct the walls.
Load-bearing masonry walls with timber-framed are common forms of construction in the
UK. If plasterboard-on-dabs is used, all the leakage paths in the house will become
interconnected which makes air sealing difficult. Lowe et al. (1994) found, however, that
when wet plastered masonry wall can potentially be several orders of magnitude more air
On the other hand, timber-framed walls are more popular in northern Europe and
North America. A study by Stephen (1998) on the BRE database found that timber framed
structures are on average tighter than masonry ones. However, after adjusting for age of
dwellings, this difference appears to be smaller. This is because most timber framed houses
in UK were recent constructions.
In a research project which goal was to give guidance in choosing appropriate materials
for air barrier system, Air-Ins Inc. (1998) tested 36 common building materials for air
leakage using laboratory test chamber experimental setup. Only half of the samples are
found to be in compliance with the Canada National Building Code limit of 0.02 l/sm
at 75
R-2000 is a program offered by Natural Resources Canada’s Office of Energy Efficiency, which
encourages and certifies the building of energy efficient houses according certain criteria.
16 OF 46
Pa. The testing found much non-homogeneity within individual sample and from one
sample to another for some of the materials.
The use of polyethylene air barrier is a common practice to reduce air leakage at walls. A
recent study by Wilcox and Weston (2001) measured the air tightness of four pairs of new
California homes built with and without spun-bonded polyolefin housewrap. The authors
found that houses with housewrap are on average 13% tighter than their counterparts. It is
expected that the impact of a housewrap air barrier would be significantly greater if the air
barrier were installed as part of a continuous pressure envelope instead of as an external
finish done in the study. Yuill and Yuill (1998) also found the technique of using housewrap
over untapped extruded polystyrene foam sheathing has the highest flow resistance among
the different materials studied. However, a longevity study by Air-Ins Inc. (1998b) showed
that spun bounded olefin paper can fail to stretch around joints under high temperature and
break away.
There are alternatives to the use of plastic film as air barrier in timber frame buildings
without sacrificing air tightness. Sikander and Olsson-Jonsson (1998) tested diffusion-
permitting polymer-based fiber sheets (sometimes known as ‘windproof’ sheets) and gypsum
board panels on three detached houses and a test structure in laboratory. Measurements
showed that it is possible to meet the Sweden Building Regulations provided if the technical
designs and quality of contractor work are of high standard. Likewise, Proskiw (1998)
concluded that both polyethylene air barrier and airtight drywall approach can meet
requirement of the Canadian R-2000 Standard based on measurements on 17 dwellings
taken over a period of eleven years. However, a study by Air-Ins Inc. (1998) found some
types of perforated polyethylene are permeable to air. After a test period of five months at
some pressure and temperature differentials, improvement in air tightness was noted due to
dust which blocked the holes.
A longevity study on the behavior of various air barrier connection techniques
submitted to pressure and temperature differentials showed that silicone base sealant and
adhesive tape are the most durable (Air-Ins Inc., 1998b). On the other hand, open cell
gaskets, mineral wool, and perforated polyethylene should not be used due to their high
permeability. Spun bonded olefin and acrylic sealant can exhibit problems at high
temperatures. There are now recommendations on specific assembly instructions for rigid air
barrier published by the Canada Mortgage and Housing Corporation (Petrone Architects,
Recent laboratory studies by Kosny et al. (1998) on insulated concrete form (ICF)
system suggested that dwellings of this sort can be more air tight than wood frame
constructions Petrie et al. (2003) tested two identical houses located side-by-side with the
only difference being one had ICF as the exterior walls and the other had conventional
wood-framed exterior walls. Air leakage measurements showed that the ICF house was 6%
to 23% less leaky than the wood-framed one, depending on the components sealed and
climate condition during the test.
17 OF 46
A few studies in Canada and the US have shown that log houses can also be quite air
tight (Buchan et al., 1996). Lateral joints were often found not to be the major leakage
source. Instead, smoke pencil tests suggested that significant leakage occurred at the corners,
the transitions between log walls and other building components, around doors and
windows, and other wall penetrations.
Apart from leakage through wall, other important components contributing to air
leakage include windows and doors, flue and fireplace, heating ducts, and the connections to
attic, basement, crawl space, and garage. Effective leakage areas of many of these building
components, including walls, are tabulated in chapter 26 of the 2001 ASHRAE
Fundamentals Handbook. About half of the data have been updated by Colliver et al. (1994)
from the 1989 version. The authors found that the best estimate values remain unchanged
with few exceptions, though the ranges of values recorded are in general much wider as the
number of data sources increased.
Window air leakage appears to be most studied and some suggested that reductions have
been successful. In Canada, a study by Henry and Patenaude (1998) tested 35 windows for
their air leakage at cold temperatures. They found that the majority of windows met or
exceeded the highest levels of air leakage performance of Canadian window standards at
normal temperatures, and many did very well even at the lowest temperatures tested. There
have also been many studies on the impaction of window air leakage on other problems such
as heat transfer (Haile et al., 1998) and condensation (Elmahdy, 2003). Desjarlais et al. (1998)
found that the air leakage of windows can be further reduced by 60% to 80% when an
additional storm window is added.
Despite so, current window testing standards do not include air leakage from the joint
between window and wall assemblies or from the sides of the windows. Louis and Nelson
(1995) presented a test methodology for quantifying this portion of air leakage.
Measurements from a few case studies show that the extraneous air leakage from window
perimeters is often higher than the air leakage through the window unit. Proskiw (1995)
showed that conventional rough-opening sealing method (i.e., packed fiber glass) can
contribute up to 14% of the total leakage of a single-family detached dwelling. This source of
air leakage can be reduced greatly by using alternative sealing method, such as casing tape,
poly-return, poly-wrap, and foamed-in-place urethane.
Dumont (1993) reports detailed measurements of air tightness revealing significant
leakage at many of the components interfaces in building. By visualization with smoke and
by reductive sealing method, Pittomvils et al. (1996) found that the connections between
wall and roof and at the top of the roof are common sources of leaks among the six low-
energy houses studied in Belgium. A solution to this problem has been addressed in a
summary report by Adalberth (1997) which provides some guidelines to practitioners on
how to achieve good air tightness. The document not only includes drawings and
specifications, but also suggests suitable materials and a quality assurance system for meeting
the goal.
18 OF 46
Research on attic-related heat and moisture flows has been underway for over a decade
in Canada. Among the first effort was quantifying the attic interface leakage areas by method
of subtraction (i.e., house ELA including the attic interface minus house ELA with attic
equally depressurized). The attic interface leakage areas were found to be fairly uniform with
an average ELA
of 330cm
among the 20 houses tested. Only tightly built R-2000 houses
had an interface leakage area of 20cm
. Wouters et al. (1997) also found insulated attics to be
a significant source of air leakage (1/3 of the total) in new Belgian dwellings.
Significant interface leakage at crawl space has also been observed. Brennan et al. (1990)
compared the ELA of the crawl space of nine dwellings against the rest of the building
envelopes and found that even with passive vents closed, crawl spaces are much leakier.
Among the 10 houses measured in British Columbia, Fugler and Moffatt (1994) found that
the interface leakage between crawl space and the rest of the house is more pronounced with
the presence of forced-air systems, instead of radiant heating. Air leakage from basement can
also bring moisture and soil contaminants into the living space. McGrath and McManus
(1996) used tracer gas techniques to measure the air flow through the basement ceiling to the
room above in two homes in UK. By visual inspection, the reason for leakiness was the
cracks between the floor-boards and between the floor and wall.
Houses built slab-on-grade or have fully conditioned basement are known to have much
less floor leakage. Sherman and Dickerhoff (1998) and Stephen (1998) observed that this
group of houses are 6% and 27% more air tight respectively than those that were built with
crawl space or have unconditioned basement. In the interest of reducing radon exposure,
sub-slab polyethylene air barriers have shown to be very effective in making concrete
basement floors airtight (Yuill et al., 2000). After proper installation, the effective leakage
area of the slab dropped to undetectable level. Buchan et al. (1992) measured the air leakage
of 13 heated basements and 1 crawl space with which preserved wood foundations were
used. Test results show that the foundations were in general tightly constructed and that
most of the air leakage occurred around the windows and headers in the basement.
Air leakage between garages and the houses have found to be significant among the 25
Canadian dwellings tested (Scanada Consultants Limited, 2001). The technique used to
measure the interface leakage area is similar to that described above for attic measurements –
the difference between depressurization of the house with the garage door opened and with
the garage simultaneously depressurized. The average ELA
is found to be 140cm
, which is
about 13% of the total air leakage. This is roughly proportional to the ratio of interface area
to house envelope area, meaning the house/garage interface is built with the same tightness
as the rest of the house envelope.
Studies by Bossaer et al. (1998) and Pittomvils et al. (1996) on Belgian dwellings also
revealed similar observations. Bossaer et al. (1998) determined the room-by-room air flow
rates at 50 Pa by means of compensating flow meter. The average garage interface air leakage
among 26 dwellings tested accounts to about 1/3 of the total leakage. Pittomvils et al. (1996)
also found that the interface between garages and the houses to be quite leaky even among
the six low-energy houses tested.
19 OF 46
Studies on the relationship between air tightness, ventilation, and energy use have
revealed the interdependency of these factors. For example, Yoshino and Zhao (1996) made
recommendations on the optimum air tightness for dwellings in using various ventilation
systems different climatic regions of Japan. Sherman and Matson (1997) estimated the
energy liability associated with providing the current levels of ventilation in US dwellings,
and found substantial energy saving by tightening building envelopes while maintaining
adequate ventilation. Zmeureanu (2000) on the other hand, found that by considering the
life-cycle energy consumption, the initial cost of renovation, and the carbon dioxide tax
credits, increase in air tightness of existing houses is not always cost-effective in the
Montreal (Canada) area.
Whole building air tightness measurements provide useful information about the energy
demand of dwellings. However, the correlation between the measured air tightness of houses
and indoor air quality is less clear. Parent et al. (1996) found that the carbon dioxide levels
measured in 30 single-family dwellings in Canada during heating season have little to do with
their respective air tightness. Bossaer et al. (1998) found the air tightness of rooms can vary
greatly in a given house, which can be part of the reason why whole building air tightness is a
poor predictor for indoor air quality.
The problem of air leakage in multi-family dwellings is more complex due to the
partition wall between units and the sheer size of the building envelope. Furthermore, there
are additional leakage pathways to be considered, e.g. adjacent units, stairwell doors, garage
chutes, elevator shafts, etc. If fan pressurization method is used, multiple blower doors
and/or very large scale equipments will be needed. Not only is the test procedure more time
and labor intensive, it also requires more cooperation from residents for accessing multiple
units simultaneously. Some of the studies discussed below used tracer gas method to
measure inter-zonal air flow. Even though the measurements themselves are not direct
measure of air tightness of the units tested, some of the findings provide insights about the
relative importance of various leakage pathways in the building.
Relative to the amount of data on single-family dwellings, there are fewer measurements
on the air leakage of multi-family dwellings. Table 2 shows some of the major studies
available from various countries. While the list is not all inclusive of past measurements, it
captures most of the recent studies on air leakage of various types of multi-family dwellings.
Measurements of air leakage of multi-family dwellings can be divided into whole
building envelope measurements, zonal measurements (floor-by-floor or unit-by-unit), and
component leakage measurements. Most data are available on unit-by-unit bases. Levin
(1991) summarized the air leakage of 53 units measured under the Stockholm Project, of
which many of them are quite air tight (0.45 to 0.9 ACH
). The air tightness of a number of
20 OF 46
apartment units in this study was measured under the condition that the adjacent units were
also pressurized. Using this method, the internal air leakage between apartment units were
found to account for 12% to 33% of the total air leakage at 50 Pa. Similar relative leakage to
internal walls has been reported by Lagus and King (1986), Reardon et al. (1987), and Love
(1990) in Canada, and Cornish (1989) in the UK, of which the test dwellings were all row
house type.
Table 2 List of recent studies on air leakage of multi-family dwellings
Country Sources Buildings # Units
Sweden Blomsterberg et al. (1995)
Levin (1991)
Boman & Lyberg (1986)
Lundin (1981)
3 Buildings
7 Buildings
3–Story Buildings
2 Terraced Houses
Canada Nichols & Gerbasi (1997)
Gulay et al. (1993)
Shaw et al. (1991)
Love (1990)
Shaw et al. (1990)
Shaw (1980)
Reardon et al. (1987)
10 Mid-Size Buildings
10 High-Rises
1 5–Story Buildings
9 Row Houses
2 High-Rises
5 High-Rises
2 Row Houses
US Lagus & King (1986)
Palmiter et al. (1995)
Flanders (1995)
Dietz et al. (1985)
Zuercher & Feustel (1983)
4 Row Houses
3–Story Buildings
3 Quadra-plexs
2 Quadra-plexs
1 High Rise
France Barles & Boulanger (2000)
Litvak et al. (2000)
3 Buildings
Multi-Family Dwellings
Russia Armstrong et al. (1996) 12 Buildings 50
Lithuania Juodis (2000) High-Rises 33
Japan Murakami & Yoshino (1983) 7 Buildings 16
UK Cornish et al. (1989) Large Panel System Dwellings 9
Finland Kovanen & Sateri (1997) 3 Buildings 8
The study used tracer gas method to measure infiltration and not air tightness directly.
21 OF 46
Borman and Lyberg (1986) analyzed 150 units from some 3-story buildings and found
that they were similar to single-family dwellings in air tightness. But such is not always the
case. Later studies by Blomsterberg et al. (1995) and Kronvall and Boman (1993) also carried
out in Sweden suggested that multi-family dwellings have lower ACH
than single-family
ones. The authors attributed this to the fact that multi-family dwellings have higher volume
to surface area ratio, and therefore lower ACH
values. Litvak et al. (2000) and Murakami
and Yoshino (1983) also observed that multi-family building units to be more air tight than
single-family ones in France and Japan respectively. Despite so, the air tightness of many
multi-family dwellings still does not meet the building code and standard in many countries.
In Canada for example, the air tightness of 10 typical mid-size buildings tested were found to
be well below the requirements of the National Building Code 1995 (Nichols and Gerbasi,
By using a multi-tracer measurement system, Palmiter et al. (1995) found significant flow
from the ground floor units directly into the top floor units in some 3-storey buildings due
to stack effect. The average flow measured in common walls with plumbing and electrical
utilities running from the ground floor to the top was larger than most of the horizontal
interzonal flows. The building tested was of standard wood frame construction, with slab-
on-grade foundation. An earlier study by Cornish (1989) in UK and Dietz et al. (1985) in US
also found similar stack induced leakage between units.
Reardon et al. (1987) found that units on the upper level were much leakier than those
below. The reason for this is because the structure was built with a concrete lower level and
wood frame upper level. Furthermore, the lower units have one less air leakage pathway –
the roof top. Vertical distribution of leakage is a concern because according to a modeling
parametric study by Sateri et al. (1995), this is the most important factor affecting infiltration.
Recent studies in countries where measurements on multi-family dwellings were not
previously available, such as in France (Barles and Boulanger, 2000) and Lithuania (Juodis,
2000), found that there is large variation in air tightness of units in a same building. At the
most extremes, 10-fold difference has been observed.
Flanders (1995) compared the air leakage of some multi-family units measured using
four fan pressurization protocols based on standards by the International Standards
Organization (ISO 9972), American Society for Testing and Materials (ASTM E779), and
Canadian General Standard Board (CAN/CGSB-149.10). The author concluded that the
three standards gave similar flow coefficient and exponent values when the weather
condition was clam, but uncertainty increases as the outdoor became windier. He
recommended that the door of the adjacent units should be left opened, instead of closed,
when carrying out blower test if the units cannot be pressurized simultaneously.
Most of the studies mentioned above are low-rise multi-family dwellings. Air leakage of
high-rise buildings has been measured in relatively large-scale study in Canada (Gulay et al.,
1993) and in Russia (Armstrong et al., 1996). Recent measurements by Barles and Boulanger
22 OF 46
(2000) in France and Juodis (2000) in Lithuania also included some high-rise residential
buildings. The Canadian study included measurements on whole building leakage, floor-by-
floor leakage, unit leakage, and component leakage. Findings confirmed that the air leakage
rates for the high-rise residential buildings far exceeded NRCC’s proposed guidelines of 0.05
to 0.15 l/sm
at 75 Pa.
Whole building air leakage test requires access to every unit and room located around
the perimeter of the building. This method requires the most cooperation from tenants and
owners. It also requires access to large-scale fan pressurization equipments. Parekh (1992)
measured two buildings before and after air sealing of the building envelope and observed
32% and 38% reduction in air leakage. The author also suggested some guidelines for
qualitative assessment of the air leakage characteristics of the building envelope by
components: windows, external doors, building envelope, elevator shafts and services shafts,
and miscellaneous including exhaust fan dampers and ducts, etc. In the summary report,
Gulay et al. (1993) tabulated the percent distribution of the whole building leakage by
component estimated based on those guidelines: 42% windows, 26% doors, 14% vertical
shafts, and 6% building envelopes.
Shaw et al. (1991, 1990, and 1980) used similar method to measure the whole building
air tightness of four high-rise apartment buildings. They found the pressure difference across
the envelope to be decreasing with building height due to large flow resistance in the
stairwell. The air flow corresponding to a height-averaged pressure difference of 50 Pa
ranged from 1.8 l/sm
to 3.6 l/sm
. The value reported by Gulay et al. (1993) which was
measured before air sealing work lied somewhere in between at 2.15 l/sm
Armstrong et al. (1996) measured the air leakage of 50 apartments located in 12
buildings and found correlation between ELA
and the apartment volume. This correlation
was particularly profound when the blower door tests were carried out with major leakage
pathways sealed, such as the windows, the balcony door, and the kitchen and bathroom
exhaust grilles. Windows and patio doors were found to contribute less than 1/3 of the total
ELA under “vents-sealed” condition. These results were, unfortunately, compromised by
variation in the incremental sealing techniques and non-uniform outside pressure on the
envelope of the tested apartment.
Leakage characteristics of stairwells have been studied by Zuercher and Feustel (1983)
on a nine-storey student dormitory. Flow coefficients and exponents were reported from the
pressurization and depressurization tests carried out under various doors/emergency doors
operation conditions. Tracer gas measurements were also carried out to study the influence
of wind and stack effect upon air infiltration.
Smoke control is another common concern in high-rise buildings. Tamura and Shaw
(1981) measured the pressure differences and flow velocities in various parts of two high-rise
buildings. Results demonstrated that the performance of the smoke shaft in venting the fire
floor can be seriously impaired by the extraneous leakage flow into the smoke shaft through
the shaft wall construction from other floors. Related studies regarding the ventilation and
23 OF 46
infiltration characteristics of lift shafts and stairwells have recently been summarized by
Limb (1998).
Shaw (1980) used an airtight test chamber to measure the leakage through windows,
walls, balcony doors, and various joints. Most of the air leakage values vary widely from
building to building, and even within the same unit. Of all the windows tested, only 1/3 of
them passed the ASHRAE 90-75 Standard. A larger fraction (2/3) of balcony doors meets
the Standard. The major air leakage sources in exterior walls are found to be floor-wall
joints, windows, and window sills.
Kovanen and Sateri (1997) measured the component leakage of two multi-family
dwellings using direct (pressure chamber) and indirect (reductive sealing) method. The main
leakage route is found, again, to be the balcony door. Three out of eight apartment units
became less air tight after renovation that was carried out without special attention on
envelope sealing, even though the air tightness of the windows and apartment doors
improved in every apartment. The most problematic component appeared to be the balcony
Measurements of the equivalent leakage areas of ten suite-access doors in some mid to
high rise apartment buildings in Canada was taken to understand their ventilation
characteristics (Wray, 2000). The leakages were found to be highly variable and did not meet
smoke control requirements, which is probably because the airflow entering the suites from
the corridor is often used as the primary ventilation air supply.
Murakami and Yoshino (1983) tested the component leakage of a few apartment units
and rooms and found there are many background leakages other than widows, doors,
ventilation inlet, and pipe openings. For example, in a bedroom tested, the leakage through
ceiling, ceiling/wall, and floor/wall joints together accounted for 3/5 of the total leakage.
Installed windows were often found to have air tightness far inferior to the performance
Exterior wall air tightness values were found to be approximately nine times greater than
those of the floor/ceiling separations in a 5-storey apartment building tested (Shaw, 1991).
Leakage to left and right partition was somewhere in between the two extremes. Good
agreement between the summations of individual leakage component and the measured
overall leakage for a unit is observed.
Juodis (2000) and Hill (2001) did not find year built to be a determining factor for the air
tightness of multi-family residential buildings. The study by Kronvall and Boman (1993),
however, found the opposite. This difference can perhaps be explained by the fact that the
later study was on Swedish dwellings where building codes have more stringent
24 OF 46
specifications on air tightness over the years. Boman and Lyberg (1986) found that if older
buildings have been retrofitted or weatherstripped, the age effect may become less
Boman and Lyberg (1986) also found that the presence of a fireplace tends to correlate
with higher air leakage in both single-family and multi-family dwellings. For dwellings that
were built between 1940 and 1960, those with fireplace have an averaged normalized leakage
area nearly twice of those without fireplace. Blomsterberg et al. (1995) found that apartments
with passive stack ventilation are much tighter than the ones with exhaust ventilation.
Shaw (1991) observed that the overall air tightness values of four buildings with
different wall constructions are not very different from each other. This is because the air
tightness value of a wall assembly is mostly dependent on how well the vapor
barrier/interior component is installed. Lundin (1981) found significant air leakage induced
by air/vapor barrier that breaks at the walls that separate apartment units. As a result,
apartment separating walls should be connected to the inside of the exterior wall to ensure a
continuous air/vapor barrier enclosing the entire wooden frame.
A recent analysis on existing air tightness data of 139 commercial and institutional
buildings by Persily (1999) found that non-residential buildings are often not air tight
enough. About half of the data analyzed were part of a study conducted by Cummings et al.
(1996) on small, predominately one-story commercial buildings. The rest include office,
industrial and retail buildings, as well as schools, from Canada, Sweden, the UK, and the US.
No correlation between air tightness and building age or wall construction was observed.
Part of the reason was that there were simply not enough data for trends to be identified.
There were some indications, however, that taller buildings tends to have more air tight
envelopes. This might be a result of more careful design and construction necessary to deal
with more demanding structural requirements, such as increased wind loads and the control
of rain penetration.
Few other measurements have been made available and they are listed in Table 3
together with those included in Persily’s analysis. Even with the new additions, air tightness
measurements of non-residential buildings remain scarce and they do not adequately
represent the existing building stock. A recent literature review by Proskiw and Phillips
(2001) summarized most of the same data as Persily’s, but with few additions of
measurements made in Canada. The bulk of their report focused on test methods and
specifications for large buildings.
25 OF 46
Table 3 List of studies on the air tightness of non-residential buildings
Country Source Buildings
Canada NRCC
University of Saskatchewan
8 Office Buildings
11 Schools
9 Supermarkets
1 Shopping Mall
1 Indoor Swimming Pool
1 Swine Building
Florida Solar Energy Center
Pennsylvania State University
8 Office Buildings
69 Small Commercial Buildings
1 Office Building
1 Library Wing
Wales School of Architecture
12 Office Buildings
12 Industrial Buildings
10 Office Buildings
6 Industrial Buildings
3 Industrial Buildings
Sweden NTRI 9 Industrial Buildings
France CETE de Lyon
2 Office Buildings
4 Schools
4 Hotels
2 Polyvalent Halls
4 Industrial Buildings
Belgium WTCE/CSTC 45 Schools
Japan KICT 3 Office Buildings
One of the earliest efforts was by Tamura and Shaw (1976) who tested eight new office
buildings in the Ottawa area. More recently, Shaw and Reardon (1995) went back to six of
these buildings which are still in use to determine the changes in their air tightness.
Comparisons indicated that as a result of various retrofit measures applied, all but two
building envelope became more air tight than 20 years ago. The improvement in the overall
air tightness value at 50 Pa ranges from 25% to 43% of its original value. The two exceptions
were one that received no retrofit measure, which deteriorated by 23% with time. The other
exception had all joints in the curtain wall recaulked in 1990, and the building showed no
change in air tightness which suggested that the retrofit measure was just sufficient to offset
the effect of aging. This study demonstrated significant improvements can be realized in the
overall air tightness by retrofit measures.
The experimental setups used by Tamura and Shaw (1976) and Shaw and Reardon
(1995) were identical, which involves pressurizing the test building using the building’s
supply air system and measuring the corresponding pressure differences across the building
envelope. In the US, Persily and Grot, (1986) tested the air tightness of seven federal
26 OF 46
buildings in a similar manner. The difference between the two test methods lay in the way
the air flow through the air-handler system was measured. While the former used a pair of
total pressure averaging tubes together with a static pressure probe to measure air flow, the
later used constant-injection tracer gas technique.
Persily and Grot (1986) also found that the federal buildings tested were comparable in
air tightness to the Canadian buildings. However, the authors commented that it was
probably more appropriate to normalize the air flow by wall area only, instead of including
roof area because the roofs were constructed to be impervious to air. Normalizing the
leakage rate with the wall area only would lead to higher values as a result.
In countries like the UK where most buildings are naturally ventilated, alternative
approach is needed. Measurements by BRE (Perera et al., 1990 and 1992) and BSRIA (Potter
et al., 1995) were obtained by attaching an external large-scale fan to the building. While the
low-energy office building tested by BRE had air tightness average of those tested in North
America, most conventional office buildings were found to be leakier by a few-folds. Litvak
et al. (2001) found that only two out of the twelve buildings sampled are in compliance with
the French RT2000 regulation. Most of the large commercial buildings tested had air
tightness in the range of that those tested in North America.
Hayakawa and Togari (1990) developed a simple test method that utilizes buoyancy
caused by the stack effect instead of using fans to pressurize test building. While the stack
effect is active, test building can be pressurized or depressurized by opening doors and
windows on the bottom floor or top floors. Under calm weather conditions, the authors
measured the equivalent effective leakage area for three high-rise office buildings. This
method had been found to be effective given if no large unknown cracks are present and the
friction resistance of the air flow in the building is small.
The study by Florida Solar Energy Center tested 69 small commercial buildings and
found that a large fraction of them were leakier than the residential homes in the area
(Cummings et al., 1996). Strip mall units were found to be 2.5 times leakier than detached
buildings. The reason for this is that the attached units were often well connected to each
other above the ceiling level.
Study by Shaw and Jones (1979) measured the air tightness of eleven Canadian schools
and found lower values than those of office buildings (Tamura and Shaw, 1976). The results
indicated that there was no meaningful relation between total energy consumption and the
measured air leakage rate. Instead, poor workmanship and sealing were observed to be the
cause of high air leakage. The air tightness of 45 Belgian schools tested by Wouters et al.
(1988) revealed a much wider range of values, even among the newly constructed schools.
The air tightness of industrial buildings has been tested by a few researchers using
similar large scale fan pressurization method. The buildings tested by Lundin (1986) in
Sweden were found to be a few folds tighter than those in the UK (Potter and Jones (1992),
Perera and Parkins (1992), Jones and Powell (1994)) and France (Fleury et al., 1998). A wider
27 OF 46
range in air tightness values were also observed in the UK and France compare to those in
Restaurants tend to have large exhaust but without enough make-up air which often
causes them to have unique uncontrolled airflow problems. Cummings et al. (1996) found
that most of the seven restaurants tested have the air barrier and the thermal barrier at
different planes, resulting in air-transported heat transfer problems. Another special type of
building tested was livestock buildings. Zhang and Barber (1995) tested the air leakage of a
new swine building and found it to be quite tight compared with office buildings also tested
in Canada (Shaw and Reardon, 1995).
Bahnfleth et al. (1999) attempted to measure the envelope air leakage of one floor of a
university library by floor-by-floor blower door method. However, the authors found that it
was impossible to adequately sealing a single floor to isolate it from its neighbors. Proskiw
and Parekh (2001) proposed a new air tightness procedure to separate the exterior envelope
air leakage from interior partition air leakage in a multi-zone building. The preliminary test
result at an indoor swimming pool which was attached to a recreational complex showed this
procedure seems to offer advantages over those of the pressure-masking technique.
To answer the need of assessing the installation of air barrier during construction period,
Knight et al. (1995) developed test equipment that is capable to handle all materials and
design configurations involved. The end product is called a Pressure Activated Chamber
Test System which used soap solution to visualize the leaks present. The authors tested the
equipment at three swimming pools, two health care facilities, and a seven-storey building
and found the test procedure to be effective in identifying leaks.
In light of the fact that many of the air leakage problems are caused by poor designs and
workmanship, practical guidelines for designers, contractors, and developers have been made
available by various agencies. For example, CMHC recommended certain jointing materials,
primary air barriers, and prefabricated assemblies that are effective in controlling air leakage
in high-rise commercial buildings (Canam Building Envelope Specialists Inc., 1999). NIST
published a document on envelope design guidelines for federal office buildings to ensure
thermal integrity and air tightness (Persily, 1993). Aside from its guidelines (Perera et al.,
1994), BRE also developed a tool for predicting the air tightness of office buildings
envelopes either at the design stage or before a major refurbishment (Perera et al., 1997).
Comparison with ten office buildings in the BRE database showed good agreement between
measurements and predictions.
Unlike residential buildings where multiple studies have suggested that new dwellings are
built tighter, Potter et al. (1995) concluded otherwise from comparison of office buildings
built before and after 1990. Similarly, Cummings et al. (1996) found that the small
commercial buildings tested did not demonstrate a clear age trend. Shaw (1981) noticed that
newly constructed supermarkets were found to be generally much leakier than the older
28 OF 46
ones, which could be explained by the opening around the receiving doors with hydraulic
No significant trend has been observed between air leakage and construction materials
of commercial buildings. However, building type can be an important factor because of the
differences in typical architecture according to their functions. For example, when compare
against hotels and schools, office buildings and polyvalent halls appear to be leakier because
of the presence of suspended ceilings (Litvak et al., 2001). The air leakages of the
supermarkets and mall tested by Shaw (1981) were also found to be higher and more spread
out than schools and high-rise office buildings measured in Canada.
Ideally air-conditioned buildings should have minimal air infiltration and naturally
ventilated buildings should have air infiltration under occupant control. By comparing
among the twelve buildings tested for air tightness, Potter et al. (1995) found that the four
naturally ventilated buildings tend to be tighter than the reminding eight which have air-
conditioning. This discloses construction practices and defects often have larger influence on
air tightness than building design.
Air leakage at suspended ceilings where electrical, lighting, and ventilation equipments
are housed has found to be significant among many of the 12 non-residential buildings
tested in France (Litvak et al., 2001). Study by the Florida Solar Energy Center (Cummings et
al., 1996) also found similarly among smaller commercial buildings. Perhaps more surprising
is that some studies from the UK have shown even the roof tops of large buildings are not
guaranteed to be impervious to air infiltration (Perera and Parkins, 1992, Potter et al., 1995).
This is somewhat counterintuitive because most would assume rain penetration problems
would have prevented any buildings from having a leaky roof top.
Cracks along the top edge of most operable windows were also found to be an
important source at a building tested, which was known to have air leakage induced
problems (Perera and Parkins, 1992). When compared to the ASHRAE window leakage
standard of 0.77 l/s/m, Persily and Grot (1986) also found that many of the windows tested
in the federal buildings exceeded that standard. However, it should be noted that the
window leakage standard exclude leakage through the window frame, which the test
procedure included besides leakage through sash.
In relative terms, Potter et al. (1995) found exposed cavities to be more problematic
than windows. This means that electrical and service penetrations through the structure into
the cavity are in need of careful sealant. Cummings et al. (1996) found this problem is
particularly disastrous among small commercial buildings where cavities are commonly used
as ducts or plenums
Duct leakage among commercial buildings is profound even after accounting for the fact
that they have greater surface area than those in residential buildings (Cummings et al.,
1996). The duct systems tested were about 70 times leakier than the SMACNA standard
29 OF 46
(Sheet Metal and Air Conditioning Contractors National Association). Depending where the
ducts are located, the impact on energy consumption can vary. However, excessive air
leakage among non-residential buildings is quite common. Among the eleven schools tested,
Shaw and Jones (1979) found that 15 to 43% of the overall air leakage can be attributed to
the air intake and exhaust openings. The leakage through roof ventilators among leaky UK
industrial buildings was found to be a bit less significant at 9% (Jones and Powell, 1994).
Despite that the test results on loading doors among UK industrial buildings were
satisfactory, Potter and Jones (1992) noticed a wide variation in the quality of the roller
shutter doors among the 12 industrial buildings tested. As improvements in the air tightness
of other parts of the building progress, this leakage component should not be neglected.
Recent work by Yuill et al. (2000b) estimated flow coefficients of automatic doors as
function of door type and rate of use.
Another common air leakage pathway is the elevator shafts as they are normally vented
to atmosphere (Potter et al., 1995). It is therefore essential for elevator doors to be fitted
with adequate seals. In an effort to insolate one floor from the others, Bahnfleth et al. (1999)
found numerous holes and cracks that could not be reached and sealed in return risers and
elevator shafts. Among other leakage components such as the stairway, a literature search by
Edwards (1999) concluded that the data on air leakage associated with elevator shafts are
very limited. Data on many other important leakage pathways, such as underground parking
garage access door and garage chutes, are even nonexistent. Nonetheless the author has
summarized some component leakage data needed to model mid- and high-rise apartment
Before concluding this state-of-the-art review it would be remiss not to mention some
of the more innovative techniques for measuring air tightness, even if they have not
generated a lot of data. The discussion so far and the vast majority of published air tightness
work is on steady-state flow. The closest most cracks and leaks actually get to steady-state
are during fan pressurization tests. In this section we will review the issues associated with
non-steady flow through relating to air tightness.
When considering time-varying air flows, there are two regimes, which we shall call
pseudo-steady state and unsteady. The difference comes about because the change in air
flow (or driving pressure) is either long or short compared to the characteristic time of the
problem at hand. The characteristic may be the time it takes sound to cross the leak or cross
the building, or it may be the time it takes a boundary layer (or jet) to form or flow the fluid
to be accelerated to steady state.
30 OF 46
In pseudo-steady state flow, the driving pressures are changing slowly enough that the
individual leaks are presumed to be instantaneously in equilibrium. Because the air leakage is
inherently non-linear, pseudo-steady state can generate complex phenomena despite the
assumption of equilibrium.
Siren (1997) has shown that turbulence can cause a 5% bias in the power law flows using
pseudo-state state assumptions due to non-linearities. Whether a 5% bias from turbulence is
acceptable will be depend on the intended use of the data. Measurements by Sharples and
Thompson (1996) confirm that there is no large difference due to these non-linearities, but
does not contain an error analysis sufficient to separate out a 5% bias from a null result.
Siren (1997) and Sharples and Thompson (1996) refer to the well-known phenomena
that occur when the flow actually begins to reverse (i.e. fluctuate). The issue of how to treat
fluctuating air flows from the perspective of ventilation is beyond our scope here, but the
physical principles of fluctuating pressures led to the development of dynamic air tightness
measurement technique knows as AC Pressurization.
Sherman and Modera
(1986) describe the physics of
AC Pressurization. The system
operates by putting a sinusoidal
volume change (of order 1 Hz)
on the inside of the building
and measuring the pressure
response. At this frequency, the
flow pseudo-steady with respect
to flow through cracks, but is
fast enough to allow
compression in the building and
thus phase shifts from which
information can be extracted.
The approach breaks down
when any individual opening
gets sufficiently large that it can
be considered unsteady. This
typically can happen for open
windows or undampered
chimneys, but not for more
normal building leaks.
AC Pressurization has no “DC” component and uses
repeated sinusoidal variations, but Sherman and Modera
(1988) have also devised a dynamic air tightness
measurement approach that has a single perturbation. Pulse
Pressurization works by providing a pressure pulse to the
inside of a building (e.g. from a compressed air tank) and
then watching the pressure decay. The power-law equation
predicts a finite recovery time for such a decay and can be
used to analyze the data to determine leakage and volume.
Like AC Pressurization the limitation of this procedure is
when unsteady flow develops. The problem for pulse
pressurization comes not from large external openings, but
from the need for the pressure to be the same throughout
the volume of the space. This tends to limit the application
to small homes or apartments, unless multiple injectors are
In a pair of papers, Dewsbury (1996, 1996a) has examined some additional analysis
approaches involving low frequencies, Fourier analysis and non-linear optimization
strategies. The lower the frequency the less susceptible the analysis is to effects of inertia
31 OF 46
and flexing of the envelope. Since low frequencies imply low pressure, signal-to-noise ratios
can become an issue.
Because of its relative complexity compared to fan pressurization, AC Pressurization has
not seen wide-spread use. It has, however, been used in some special circumstances when
fan pressurization was undesirable.
The physics of air leakage through building components is non-linear. The non-linearity
of the process can lead to some challenging measurement and interpretation problems. The
fundamental form of the air leakage equations are not a priori clear, but there is general
agreement that a power-law formulation is theoretically justifiable and empirically valid.
There is less consensus on how to report air leakage data and several metrics are
commonly in use. The difference of opinion comes in part from the fact that different
quantities are useful for different purposes. Assuming a power-law description, all two-
parameter (unnormalized) formalizations are interchangeable. Single parameter forms
provide less accuracy, but can be useful for specific purposes.
Regardless of the parameterization chosen air leakage data shows a huge scatter even
within ostensibly homogeneous populations. It is not atypical to see log-normal
distributions with the standard deviation being equal to the mean. The large variation can be
attributable to variations in workmanship, variations in structure use and maintenance and
variations in renovation and repair activities.
Despite the variance there are some very general and not overly surprising trends
that can be teased from the data. The air leakage characteristics of single-family dwellings are
better understood than multi-family dwellings or non-residential buildings because more
measurements are available. Dwellings in more severe climate, like those in Sweden and
Canada, have shown to be more air tight than those in the US and the UK, where the
climate is milder. In countries where there is a demand for tighter envelopes driven by
building codes or energy savings, new constructions has been shown to more air tight than
older ones. Dwellings of different construction types have different envelope air tightness
properties, but some air leakage pathways are common among many dwellings, such as the
connections between building materials and components. Leakage to attics, basements, crawl
spaces, and garages is significant and raises addition energy and health concerns. Many
studies have addressed the effectiveness of air barriers and building materials to minimize
leakage, but it is often the quality of workmanship and careful design that are the
determining factors in achieving desirable air tightness.
When compared to single-family dwellings, individual units in multi-family dwellings
tend to be more air tight. However, this does not mean that multi-family buildings are
sufficiently air tight, particularly for the high-rise buildings. Despite that the air leakage to the
exteriors still dominates, studies have also revealed significant air leakage between units in
32 OF 46
multi-family dwellings. Stack induced vertical air flow between units and in elevator shafts
and stairwells are among some of the concerns. Partly limited by the number of
measurements available, few trends have been observed between building characteristics and
air tightness. The task of identifying air leakage trends is further complicated by large
variations in air tightness found between units in a same building. Many of the findings
observed among single-family dwellings also apply to multi-family dwellings, such as:
dwellings with fireplaces tend to be leakier, and the integrity of the air barrier system is
crucial to ensure air tightness of the unit.
Office buildings, industrial buildings, schools, and retail stores are among the few
non-residential building types of which air tightness measurements are available. As
measurements in these buildings often required large scale equipment, a few alternative
methods have been proposed such that measurements can be made more easily and less
costly. However, the applications of these methods remain research-grade. In fact, the most
recent measurements were collected using large-scale fan pressurization almost exclusively. It
is evident that commercial buildings are rarely air tight enough. There is a slight geographical
difference in the air tightness of buildings in Sweden (most tight), the UK (most leaky), and
the North America (somewhere in between). On the other hand, air tightness is unrelated to
age or construction materials. Suspended ceilings, exposed cavities, and ventilation ducts are
among the key leakage pathways. Due to the architectural differences of different building
types, some tend to be leakier than the others. But until more data have been collected, these
trends remain scattered observations that cannot be generalize to various commercial
building type. To provide more immediate help to designers and contractors, various
organizations have recently published practical guidelines to effectively control air leakage in
commercial buildings.
33 OF 46
A Area [m
Discharge coefficient [-]
C Power-law coefficient [m
d Diameter of pipe [m]
l Length (along flow path) of pipe [m]
m Mass flow correction [2.28]
n Power-law exponent [-]
Q Air flow [m
Re Reynolds number [-]
S S Number [-]
P Pressure drop [Pa]
µ Viscosity of fluid [kg/m-s]
ν Kinematic viscosity of fluid (µ⁄ρ)
φ Exponential form factor [-]
ρ Density [kg/m
This work was supported by the Assistant Secretary for Energy Efficiency and
Renewable Energy, Building Technology Program of the U.S. Department of Energy, under
Contract no. DE-AC03-76SF00098
Adalberth K., "Airtight Building - A Practical Guide," in Proceedings: Ventilation and
Cooling, 18th Air Infiltration and Ventilation Centre Conference, Athens, Greece, September 23-26,
Air-Ins Inc., "Air Permeance of Building Materials”, Research Highlights, Technical
Series 98-108, Canada Mortgage and Housing Corporation, 1998.
Air-Ins Inc., "Airtightness Tests on Components Used to Join Different or Similar
Materials of the Building Envelope”, Research Highlights, Technical Series 98-121, Canada
Mortgage and Housing Corporation, 1998b.
Armstrong P., Dirks J., Klevgard L., Matrosov Y., Olkinuora J. and Saum D.,
"Infiltration and Ventilation in Russian Multi-Family Buildings", in Proceedings: Profiting from
34 OF 46
Energy Efficiency, Summer Study on Energy Efficiency in Buildings, American Council for an Energy
Efficient Economy, Washington DC, 1996.
ASHRAE Handbook of Fundamentals, Ch 26, American Society of Heating,
Refrigerating and Air conditioning Engineers, 2001.
ASHRAE Standard 119, “Air Leakage Performance for Detached Single-Family
Residential Buildings”, American Society of Heating, Refrigerating and Air conditioning
Engineers, 1988.
ASTM, E1554, “Standard Test Method for Determining External Air Leakage of Air
Distribution Systems”, ASTM Book of Standards, American Society of Testing and
Materials, Vol. 4 (11), 2004.
ASTM, Standard E1827-96, “Standard Test Methods for Determining Airtightness of
Buildings Using an Orifice Blower Door”, ASTM Book of Standards, American Society of
Testing and Materials, Vol. 4 (11), 2002.
ASTM, Standard E779-99, “Test Method for Determining Air Leakage by Fan
Pressurization”, ASTM Book of Standards, American Society of Testing and Materials, Vol.
4 (11), 2000.
Bahnfleth W.P., G.K. Yuill. and Lee B.W., “Protocol for Field Testing of Tall Buildings
to Determine Envelope Air Leakage Rate”, ASHRAE Transactions, Vol. 105 (2), 1999, pp.
Barles, P. and Boulanger X., “Airtightness and Underpressures Measurements in French
Apartments”, in Proceedings: Innovations in Ventilation Technology, 21
Air Infiltration and
Ventilation Centre Conference, Hague, Netherlands, September 26-29, 2000.
Bassett M., "The Infiltration Component of Ventilation in New Zealand Houses”, in
proceedings: Ventilation Strategies and Measurement Techniques, 6
Air Infiltration and Ventilation
Centre Conference, Netherlands, September 16-19, 1985.
Blomsterberg A., “Air Leakage in Dwellings”, Dept. Bldg. Constr. Report No. 15,
Swedish Royal Institute of Technology, 1977.
Blomsterberg A., Carlsson T. and Kronvall J., “Short Term and Long Term
Measurements of Ventilation in Dwellings”, in Proceedings: Implementing the Results of Ventilation
Research, 16
Air Infiltration and Ventilation Centre Conference, Palm Springs, USA, September 19-
22, 1995.
Borman C.A. and Lyberg M.D., "Analysis of Air Change Rates in Swedish Residential
Buildings", Measured Air Leakage of Buildings, ASTM STP 904, Trechsel H.R. and Lagus P.L.,
Ed., American Society for Testing and Materials, Philadelphia, 1986, pp.399-406.
35 OF 46
Bossaer A., Demeester J., Wouters P., Vandermarke B. and Vangroenweghe W.,
"Airtightness Performance in New Belgian Dwellings”, in Proceedings: Ventilation Technologies in
Urban Areas, 19
Air Infiltration and Ventilation Centre Conference, Oslo, Norway, September 28-
30, 1998.
Brennan T., Pyle B., Williamson A., Balzer F. and Osborne M., "Fan Door Testing on
Crawl Space Buildings", Air Change Rate and Airtightness in Buildings, ASTM STP 1067,
Sherman M.H., Ed., American Society of Testing and Materials, Philadelphia, 1990, pp.146-
Buchan, Lawton, Parent Limited, “Airtightness and Air Quality in Preserved Wood
Foundations”, Final Report, Canada Mortgage and Housing Corporation, 1992.
Buchan, Lawton, Parent Limited, Jools Development, and Drerup Armstrong Limited.,
“Air Leakage Performance of 11 Log Houses in Eastern Ontario & Western Quebec”,
Research Report, Canada Mortgage and Housing Corporation, 1996.
Caffey G.E., “Residential Air Infiltration”, ASHRAE Transactions, Vol. 85 (9), 1979, pp.
CAN/CGSB Standard 149, “Determination of the Airtightness of Building Envelopes
by Fan Depressurization Method”, Canadian General Standards Board, 1986.
Canam Building Envelope Specialists Inc., “Practical Guidelines for Designers,
Contractors, and Developers on the Installation of Air Leakage Control Measures in New
and Existing High-Rise Commercial Buildings”, Research Highlights Technical Series 99-
119, Canada Mortgage and Housing Corporation, 1999.
Carrie F.R., Wouters P., Ducarne D., Andersson J., Faysse J.C., Chaffois P., Kilberger
M. and Patriarca V., “Impacts of air distribution system leakage in Europe: the SAVE-
DUCT project,” in Proceedings: Ventilation and Cooling, 18th AIVC Conference, Athens, Greece,
September 23-26, 1997, pp. 23-26.
Chan W.R., Price P.N., Sohn M.D. and Gadgil A.J. "Analysis of U.S. Residential Houses
Air Leakage Database”, Report 53367, Lawrence Berkeley National Laboratory, 2003.
Colliver D.G., Murphy W.E. and Sun W., "Development of a Building Component Air
Leakage Data Base”, ASHRAE Transactions, Vol. 100 (1), 1994, pp. 292-305.
Cornish J.P., Henderson G., Uglow C.E., Stephen R.K., Southern J.R. and Sanders C.H.,
“Improving the Habitability of Large Panel System Dwellings”, Report 154, Building
Research Establishment, United Kingdom, 1989.
Cummings J.B., Withers C.R., Moyer N.A., Fairey P.W. and McKendry B.B.,
“Uncontrolled Air Flow in Non-Residential Buildings”, Final Report, FSEC-CR-878-96,
Florida Energy Solar Center, 1996.
36 OF 46
Desjarlais A.O., Childs K.W., Christian J.E., “To Storm or Not to Storm: Measurement
Method to Quantity Impact of Exterior Envelope Airtightness on Energy Usage Prior to
Construction”, in Proceedings: Thermal Performance of Exterior Envelopes of Buildings VII,
Clearwater Bay, Florida, December 6-10, 1998, pp. 607-618.
Dewsbury J., “AC pressurization: Analysis of non-linear optimization”, Building Services
Engineering Research & Technology, Vol. 17 (2), 1996, pp. 65-71.
Dewsbury J., “AC pressurization: Fourier analysis and the effect of compressibility”,
Building Services Engineering Research & Technology, Vol. 17 (2), 1996b, pp. 73-77.
Diamond R.C., Dickinson J.B., Lipschutz R.D., O’Regan B. and Schole B., “The House
Doctor’s Manual”, Report 3017, Lawrence Berkeley National Laboratory, 1982.
Dietz R.N., D’Ottavio T.W. and Goodrich R.W., “Multizone Infiltration Measurements
in Homes and Buildings Using a Passive Perluorocarbon Tracer Method”, ASHRAE
Transactions, Vol. 91 (2), 1985, pp. 1761-1776.
Dumont R.S., "An Overview of Insulation and Airtightness in Wood Frame
Construction”, in Proceedings: Innovative Housing, Canada National Research Council,
Vancouver, Canada, June 21-25, 1993.
Edwards C., “Modeling of Ventilation and Infiltration Energy Impacts in Mid and High-
Rise Apartment Buildings”, Report, Sheltair Scientific Limited, 1999.
Elmahdy A.H., "Quantification of Air Leakage Effects on the Condensation Resistance
of Windows”, ASHRAE Transactions, Vol. 109 (1), 2003, pp. 600-606.
Etheridge, D.W., "Crack Flow Equations and Scale Effect", Building and Environment, Vol.
12, 1977, pp. 181-189.
Flanders S.N., “Fan Pressurization Measurements by Four Protocols”, in Proceedings:
Implementing the Results of Ventilation Research, 16
Air Infiltration and Ventilation Centre Conference,
Palm Springs, September 19-22, 1995.
Fleury E., Millet J.R., Villenave J.G., Veyrat O. and Morisseau C., “Theoretical and Field
Study of Air Change in Industrial Buildings”, in Proceedings: Ventilation Technologies in Urban
Areas, 19
Air Infiltration and Ventilation Centre Conference, Oslo, Norway, September 28-30,
1998, pp. 57-65.
Francisco P., “The current state of duct leakage measurements: field evaluation of five
techniques,” Home Energy, March/April 2001, pp. 33-35.
Fugler D.W. and Moffatt S.D., "Investigation of Crawl Space Performance in British
Columbia”, ASHRAE Transactions, Vol. 100 (1), 1994, pp. 1411-1419.
37 OF 46
Fugler D.W., “Conclusions from Ten Years of Canadian Attic Research”, ASHRAE
Transactions, Vol. 105 (1), 1999, pp. 819-825.
Gulay B.W., Stewart C.D. and Foley G.J., “Field Investigation Survey of Airtightness,
Air Movement and Indoor Air Quality in High Rise Apartment Buildings Summary Report”,
Canada Mortgage and Housing Corporation, 1993.
Gusdorf J., Personal communication, October 2003. Also see EnerGuide for Houses
Program, Natural Resources Canada:
Haile S., Bernier M.A., Patenaude A. and Jutras R., "The Combined Effect of Air
Leakage and Conductive Heat Transfer in Window Frames and its Impacts on the Canadian
Energy Rating Procedure”, ASHRAE Transactions, Vol. 104 (1a), 1998, pp.176-184.
Hamlin T and Gusdorf J., “Airtightness and Energy Efficiency of New Conventional
and R-2000 Housing in Canada, 1997”, The CANMET Energy Technology Centre,
Department of Natural Resources Canada, 1997.
Harrje D.T., Blomsterberg A. and Persily A.K., “Reduction of Air Infiltration Due to
Window and Door Retrofits”, CU/CEES Report 85, Princeton University, 1979.
Harrje D.T., Dutt G.S. and Beya J.E., “Locating and Eliminating Obscure, but Major
Energy Losses in Residential Housing,” ASHRAE Transactions, Vol. 85 (2), 1979, pp. 521-
Harrje D.T., “Building Envelope Performance Testing,” ASHRAE Journal, March 1981,
pp. 39-41.
Harrje D.T. and Dutt G.S., “House Doctors Program: Retrofits in Existing Buildings,”
in Proceedings: 2
Air Infiltration and Ventilation Centre Conference, 1981, pp. 61-72.
Hayakawa S. and Togari S., “Simple Test Method for Evaluating Exterior Wall
Airtightness of Tall Office Buildings”, Air Change Rate and Airtightness in Buildings, ASTM STP
1067, Sherman M.H., Ed., American Society for Testing and Materials, Philadelphia, 1990,
pp. 231-245.
Henry R. and Patenaude A., "Measurements of Window Air Leakage at Cold
Temperatures and Impact on Annual Energy Performance of a House”, ASHRAE
Transactions, Vol. 104 (1b), 1998, pp. 1254-1260.
Hill D., “Valuing Air Barrier”, Home Energy, September/October, 2001, pp. 29-32.
ISO, Standard 9972, “Thermal Insulation-Determination of Building Air Tightness –
Fan Pressurization Method”, International Standards Organization, 1996.
38 OF 46
Jones P.J. and Powell G., “Reducing Air Infiltration Losses in Naturally Ventilated
Industrial Buildings”, in Proceedings: The Role of Ventilation, 15
Air Infiltration and Ventilation
Centre Conference, Buxton, United Kingdom, September 27-30, 1994, pp. 397-409.
Juodis E., “Energy Saving and Airtightness of Blocks of Flats in Lithuania”, Indoor Built
Environment, Vol. 9, 2000, pp. 143-147.
Knight K., Knight G., Sharp J., Guerriero J. and Phillips B., “Evaluating Test
Equipment for Air Tightness of Construction Details”, Report, Canada Mortgage and
Housing Corporation, 1995.
Kosny J., Christian J.E., Desjarlais A.O., “Performance Check between Whole Building
Thermal Performance Criteria and Exterior Wall Measured Clear Wall R-Value, Thermal
Bridging, Thermal Mass, and Airtightness”, ASHRAE Transactions, Vol. 104 (2), 1998, pp.
Kronvall J., “Air Tightness Measurements and Measurement Methods”, Report D8,
Swedish Council for Building Research, Stockholm, 1980.
Kronvall J. and Boman C.A., "Ventilation Rates and Air Tightness Levels in the Swedish
Housing Stock", in Proceedings: Energy Impact of Ventilation and Air Infiltration, 14
Air Infiltration
and Ventilation Centre Conference, Copenhagen, Denmark, September 21-23, 1993.
Kovanen K. and Sateri J., “Airtightness of Apartments Before and After Renovation”, in
Proceedings: Ventilation and Cooling, 18
Air Infiltration and Ventilation Centre Conference, Athens,
Greece, September 23-26, 1997, pp. 117-125.
Kula H.G. and Sharples S., “Air Flow through Smoot and Rough Cracks”, in Proceedings:
The Role of Ventilation, 15
Air Infiltration and Ventilation Centre Conference, Buxton, United
Kingdom, September 27-30, 1994, pp. 710-717.
Lagus P.L. and King J.C., “Air Leakage and Fan Pressurization Measurements in
Selected Naval Housing”, Measured Air Leakage of Buildings, ASTM STP 904, Trechsel H.R.
and Lagus P.L., Eds., American Society for Testing and Materials, Philadelphia, 1986, pp. 5-
Levin P., "Building Technology and Air Flow Control in Housing", Report D16,
Swedish Council for Building Research, Stockholm, 1991.
Limb M.J., “Ventilation and Infiltration Characteristics of Life Shafts and Stair Wells”,
Annotated Bibliography 4, Air Infiltration and Ventilation Centre, Coventry, 1998.
Litvak A., Kilberger M. and Guillot K., “Field Measurement Results of the Airtightness
of 64 French Dwellings”, in Proceedings: Air Distribution in Rooms, Roomvent, 2000, pp. 1093-
39 OF 46
Litvak A., Boze D. and Kilberger M., “Airtightness of 12 Non Residential Large
Buildings Results from Field Measurement Studies”, in Proceedings: Market Opportunities for
Advanced Ventilation Technology, 22
Air Infiltration and Ventilation Centre Conference, Bath, United
Kingdom, September 11-14, 2001.
Louis M.J. and Nelson P.E., "Extraneous Air Leakage from Window Perimeters",
Airflow Performance of Building Envelopes, Components, and Systems, ASTM STP 1255, Modera
M.P. and Persily A.K., Eds., American Society for Testing and Materials, Philadelphia, 1995,
Love J.A., "Airtightness Survey of Row Houses in Calgary, Alberta", Air Change Rate and
Airtightness in Buildings, ASTM STP 1067, Sherman M.H., Ed., American Society for Testing
and Materials, Philadelphia, 1990, pp.194-210.
Lowe R.J., Curwell S.R., Bell M. and Ahmad A., "Airtightness in Masonry Dwellings:
Laboratory and Field Experience”, Building Services Engineers Research Technology, Vol. 15
(3), 1994, pp.149-155.
Lowe R.J., Johnston D. and Bell M., "Airtightness in UK Dwellings: A Review of Some
Recent Measurements”, in Proceedings: The Second International Conference on Buildings and the
Environment, Paris, June 9-12, 1997.
Lundin L.I., “Airtightness in Terraced Houses”, in Proceedings: Building Design for Minimum
Air Infiltration, 2
Air Infiltration and Ventilation Centre Conference, Sweden, September 21-23,
1981, pp. 185-195.
Lundin L.I., “Air Leakage in Industrial Buildings – Description of Equipment”, Measured
Air Leakage of Buildings, ASTM STP 904, Trechsel H.R. and Lagus P.L., Eds., American
Society for Testing and Materials, Philadelphia, 1986, pp. 101-105.
Matson N.E., Feustel H.E., Warner J.L. and Talbott J., "Climate-Based Analysis of
Residential Ventilation Options: New York Analysis”, in Proceedings: The Role of Ventilation,
Air Infiltration and Ventilation Centre Conference, Buxton, United Kingdom, September 27-
30, 1994.
McGrath P.T. and McManus J., "Air Infiltration from Basements and Sub-floors to the
Living Space”, Building Services Engineers Research Technology, Vol. 17 (2), 1996, pp. 85-87.
McWilliams J.A., "Review of Airflow Measurement Techniques'', Annotated
Bibliography 12, Air Infiltration and Ventilation Centre, 2002.
Modera M.P. and Wilson D.J., “The Effects of Wind on Residential Building Leakage
Measurements”, Air Change Rate and Airtightness in Buildings, ASTM STP 1067, Sherman M.H.,
Ed., American Society for Testing and Materials, Philadelphia, 1990, pp.132-144.
40 OF 46
Murakami S., Yoshino H., “Air-Tightness of Residential Buildings in Japan”, in
Proceedings: Air Infiltration Reduction in Existing Buildings, 4
Air Infiltration and Ventilation Centre
Conference, Elm, Switzerland, September 26-28, 1983.
Nichols L. and Gerbasi D., "Field Investigation into Airtightness, Air Movement and
Quality of Inside Air in Medium-Sized Residential Buildings and an Energy Audit of These
Buildings," Research Highlights, Technical Series 97-106, Canada Mortgage and Housing
Corporation, 1997.
Orme M., Liddament M. and Wilson A., "An Analysis and Data Summary of the AIVC's
Numerical Database”, Technical Note 44, Air Infiltration and Ventilation Centre, Coventry,
Palmiter L., Heller J. and Sherman M., "Measured Airflow in a Multifamily Building",
Airflow Performance of Building Envelopes, Components, and Systems, ASTM STP 1255, Modera
M.P. and Persily A.K., Ed., American Society for Testing and Materials, Philadelphia, 1995,
Parekh A., “Power Demand and Energy Savings through Air Leakage Control in High-
Rise Residential Buildings in Cold Climates”, in Proceedings: Thermal Performance of the Exterior
Envelopes of Buildings V, Clearwater Beach, Florida, December 7-10, 1992, pp. 632-642.
Parent D., Stricker S. and Fugler D., "Ventilation in Houses with Distributed Heating
Systems”, in Proceedings: Optimum Ventilation and Air Flow Control in Buildings, 17
Air Infiltration
and Ventilation Centre Conference, Gothenburg, Sweden, September 17-20, 1996, pp. 186-195.
Perera M.D.A.E.S., Stephan R.K. and Tull R.G., “Airtightness Measurements in Two
UK Office Buildings”, Air Change Rate and Airtightness in Buildings, ASTM STP 1067, Sherman
M.H., Ed., American Society for Testing and Materials, Philadelphia, 1990, pp. 211-221.
Perera M.D.A.E.S. and Parkins L., “Airtightness of US Buildings: Status and Future
Possibilities”, Environmental Policy and Practice, Vol. 2 (2), 1992, pp. 143-160.
Perera M.D.A.E.S., Turner C.H.C. and Scivyer C.R.., “Minimizing Air Infiltration in
Office Buildings”, Report 265, Building Research Establishment, United Kingdom, 1994.
Perera M.D.A.E.S., Henderson J. and Webb B.C., “Predicting Envelope Air Leakage in
Large Commercial Buildings Before Construction”, in Proceedings: Ventilation and Cooling, 18
Air Infiltration and Ventilation Centre Conference, Athens, Greece, September 23-26, 1997, pp.
Persily A.K., "Measurements of Air Infiltration and Airtightness in Passive Solar
Homes”, Measured Air Leakage of Buildings, ASTM STP 904, Trechsel H.R. and Lagus P.L.,
Eds., American Society for Testing and Materials, Philadelphia, 1986, pp.46-60.
41 OF 46
Persily A.K. and Grot R.A., “Pressurization Testing of Federal Buildings”, Measured Air
Leakage of Buildings, ASTM STP 904, Trechsel T.R. and Lagus P.L, Eds., American Society
for Testing and Materials, Philadelphia, 1986, pp. 184-200.
Persily A.K., “Envelope Design Guidelines for Federal Office Buildings: Thermal
Integrity and Airtightness”, National Institute of Standards and Technology, USA, 1993.
Persily A.K., “Myths About Building Envelopes”, ASHRAE Journal, Vol. 41 (3), 1999,
pp. 39-47.
Petrie T.W., Kosny J., Desjarlais A.O., Atchley J.A., Childs P.W., Ternes M.P. and
Christian J.E., “How Insulating Concrete Form vs. Conventional Construction of Exterior
Walls Affects Whole Building Energy Consumption: Results from a Field Study and
Simulation of Side-by-side Houses”, Report, Oak Ridge National Laboratory, 2003.
Petrone Architects, "Rigid Air Barrier Assemblies”, Research Highlights, Technical
Series 00-111, Canada Mortgage and Housing Corporation, 2000.
Pittomvils J., Hens H. and Van Bael F., "Evaluation of Ventilation System in Very Low
Energy Houses”, in Proceedings: Optimum Ventilation and Air Flow Control in Buildings, 17
Infiltration and Ventilation Centre Conference, Gothenburg, Sweden, September 17-20, 1996.
Potter I.N. and Jones T.J., “Ventilation Heat Loss in Factories and Warehouses”,
Technical Note TN 7/92, Building Services Research and Information Association, 1992.
Potter I.N., Jones T.J. and Booth W.B., “Air Leakage of Office Buildings”, Technical
Note TN 8/95, Building Services Research and Information Association, 1995.
Proskiw G., "Air Leakage Characteristics of Various Rough-Opening Sealing Methods
for Windows and Doors”, Airflow Performance of Building Envelopes, Components, and Systems,
ASTM STP 1255, Modera M.P. and Persily A.K., Eds., American Society for Testing and
Materials, Philadelphia, 1995, pp.123-134.
Proskiw G., "Measured Airtightness of 24 Detached Houses over Periods of up to
Three Years", Airflow Performance of Building Envelopes, Components, and Systems, ASTM STP
1255, Modera M. P. and Persily A. K., American Society for Testing and Materials,
Philadelphia, 1995b, pp. 248-265.
Proskiw G., "The Variation of Airtightness of Wood-Frame Houses over an 11-Year
Period", in Proceedings: Thermal Performance of The Exterior Envelopes of Buildings VII, Clearwater
Beach, Florida, December 6-10, 1998, pp. 745-751.
Proskiw G. and Phillips B., “Air leakage Characteristics, Test Methods, and
Specifications for Large Buildings”, Report, Canada Mortgage and Housing Corporation,
42 OF 46
Proskiw G. and Parekh A., “A Proposed Test Procedure for Separating Exterior
Envelope Air Leakage from Interior Partition Air Leakage”, in Proceedings: Performance of
Exterior Envelopes of Whole Buildings VIII, Integration of Building Envelopes, Clearwater
Beach, Florida, December 2-7, 2001.
Reardon J.T., Kim A.K. and Shaw C.Y., “Balanced Fan Depressurization Method for
Measuring Component and Overall Air Leakage in Single- and Multifamily Dwellings”,
ASHRAE Transactions, Vol. 93 (2), 1987, pp. 137-152.
Sateri J., Heikkinen J. and Pallari M., "Feasibility of Ventilation Heat Recovery in
Retrofitting Multi-Family Buildings", in Proceedings: Implementing the Results of Ventilation
Research, 16
Air Infiltration and Ventilation Centre Conference, Palm Springs, USA, September 19-
22, 1995.
Scanada Consultants Limited, "Air Infiltration from Attached Garages in Canadian
Houses”, Research Highlights, Technical Series 01-122, Canada Mortgage and Housing
Corporation, 2001
Sharples S. and Thompson D., “Experimental Study of Crack Flow with Varying
Pressure Differentials,” in Proceedings: Optimum Ventilation and Air Flow Control in Buildings, 17
Air Infiltration and Ventilation Centre Conference, Gothenburg, Sweden, September 17-20, 1996,
pp. 243-253.
Shaw C.Y. and Jones L., “Air Tightness and Air Infiltration of School Buildings”,
ASHRAE Transactions, Vol. 85 (1), 1979, pp. 85-95.
Shaw C.Y., “Methods for Conducting Small-Scale Pressurization Tests and Air Leakage
Data of Multi-Storey Apartment Buildings”, ASHRAE Transactions, Vol. 86 (1), 1980, pp.
Shaw C.Y., “Air Tightness: Supermarkets and Shopping Malls”, ASHRAE Journal,
March 1981, pp. 44-46.
Shaw C.Y., Gasparetto S. and Reardon J.T., “Methods for Measuring Air Leakage in
High-Rise Apartments”, Air Change Rate and Airtightness in Buildings, ASTM STP 1067,
Sherman M.H., Ed., American Society for Testing and Materials, Philadelphia, 1990, pp. 222-
Shaw C.Y., Magee R.J., Rousseau J., “Overall and Component Airtightness Values of a
Five-Story Apartment Building”, ASHRAE Transactions, Vol. 97 (2), 1991, pp. 347-353.
Shaw C.Y. and Reardon J.T., “Changes in Airtightness Levels of Six Office Buildings”,
Airflow Performance of Building Envelopes, Components, and Systems, ASTM STP 1255, Modera
M.P. and Persily A.K., Eds., American Society for Testing and Materials, Philadelphia, 1995,
43 OF 46
Sherman M.H. and Modera M.P., “Low Frequency Measurement of the Leakage in
Enclosures”, Review of Scientific Instruments, Vol. 57 (7), 1986, pp. 1427-1430.
Sherman M.H., “Estimation of Infiltration from Leakage and Climate Indicators”, Energy
and Buildings, Vol. 10, 1988, pp. 81-86.
Sherman M.H. and Modera M.P., “Signal Attenuation Due to Cavity Leakage”, Journal of
the Acoustic Society of America, Vol. 84 (6), 1988, pp. 2163-2169.
Sherman M.H., "A Power Law Formulation of Laminar Flow in Short Pipes", Journal of
Fluids Engineering, Vol. 114, 1992, pp. 601-605.
Sherman M.H. and Palmiter L., "Uncertainty in Fan Pressurization Measurements'',
Airflow Performance of Building Envelopes, Components, and Systems, ASTM STP 1255, Modera
M.P. and Persily A.K., Eds., American Society for Testing and Materials, Philadelphia, 1995,
pp. 266-283.
Sherman M.H. and Matson N.E., “Residential Ventilation and Energy Characteristics”,
ASHRAE Transactions, Vol.103 (1), 1997, pp. 717-730.
Sherman M.H. and Dickerhoff D., “Air-Tightness of U.S. Dwellings”, ASHRAE
Transactions, Vol. 104 (2), 1998, pp. 1359-1367.
Sherman M.H. and Matson N.E., "Air Tightness of New Houses in the U.S.", in
Proceedings: Market Opportunities for Advanced Ventilation Technology, 22
Air Infiltration and
Ventilation Centre Conference, Bath, United Kingdom, September 11-14, 2001.
Sikander E. and Olsson-Jonsson A., "Airtightness of Timber Frame Buildings Not
Having a Plastic Film Vapor Barrier”, in Proceedings: Ventilation Technologies in Urban Areas, 19
Air Infiltration and Ventilation Centre Conference, Oslo, Norway, September 28-30, 1998.
Siren K., “A Modification of the Power Law Equation to Account for Large-Scale Wind
Turbulence”, in Proceedings: Ventilation and Cooling, 18
Air Infiltration and Ventilation Centre
Conference, Athens, Greece, September 23-26, 1997, pp. 557-561.
Stephen R.K., "Airtightness in UK Dwellings: BRE's Test Results and Their
Significance”, Report 359, Building Research Establishment, United Kingdom, 1998.
Tamura G.T. and Shaw C.Y., “Studies on Extorter Wall Air Tightness and Air
Infiltration of Tall Buildings”, ASHRAE Transactions, Vol. 82 (2), 1976, pp. 122-134.
Tamura G.T. and Shaw C.Y., “Field Checks on Building Pressurization for Smoke
Control in High-Rise Buildings”, ASHRAE Journal, February, 1981, pp. 21-25.
44 OF 46
Walker I.S., Wilson D.J. and Sherman M.H., "A Comparison of the Power Law to
Quadratic Formulations for Air Infiltration Calculations", Energy and Buildings, June 1997, pp.
Wray C.P., “Suite Ventilation Characteristics of Current Canadian Mid- and High-Rise
Residential Buildings”, Report 43254, Lawrence Berkeley National Laboratory, 2000.
Wilcox B.A. and Weston T.A., "Measured Infiltration Reduction in California
Production Houses Using Housewrap”, in Proceedings: Performance of Exterior Envelopes of Whole
Buildings VIII, Integration of Building Envelopes, Clearwater Beach, Florida, December 2-7, 2001.
Wouters P., L’Heureux D., Voordecker P. and Bossicard R., “Ventilation and Air
Quality in Belgian Buildings: A State of the Art”, in Proceedings: Effective Ventilation, 9
Infiltration and Ventilation Centre Conference, Gent Belgium, September 12-15, 1988, Paper 23.
Wouters P., Bossaer A., Demeester J., Ducarme D., Vandermarke B. and
Vangroenweghe W., "Airtightness of New Belgian Dwellings: An Overview Picture”, in
Proceedings: Ventilation and Cooling, 18
Air Infiltration and Ventilation Centre Conference, Athens,
Greece, September 23-26, 1997.
Yoshino H. and Zhao Y., "Numerical Study of Building Airtightness for Optimum
Indoor Air Quality and Energy Efficiency”, in Proceedings: 5
International Conference on Air
Distribution in Rooms, Yokohama, Japan, July 17-19, 1996, Vol. 2, pp. 23-30.
Yuill G.K., Yuill D.P., “Development of a Field Procedure to Measure the Airtightness
of Wall Construction Elements of Houses”, in Proceedings: Thermal Performance of Exterior
Envelopes of Buildings VII, Clearwater Bay, Florida, December 6-10, 1998, pp. 753-765.
Yuill G.K. & Associates (Man.) Limited, “Airtightness of Concrete Basement Slabs”,
Research Highlights, Technical Series 00-130, Canada Mortgage and Housing Corporation,
Yuill G.K., Upham R. and Hui C., “Air Leakage through Automatic Doors”, ASHRAE
Transactions, Vol. 106 (2), 2000b, pp. 145-160.
Zeller J. and Werner J., "Airtightness of buildings, measurements on houses with a very
low heating energy consumption”, in Proceedings: International Symposium Energy Efficient
Buildings: Design, Performance, and Operation - CIB Working Commission W67 "Energy Conservation
in the Built Environment" and IEA-SHC Working Group Task XIII "Low Energy Buildings",
Leinfelden-Echterdingen, Germany, March 9-11, 1993.
Zhang Y. and Barber E.M., “Infiltration Rates for a New Swine Building”, ASHRAE
Transactions, Vol. 101 (1), 1995, pp. 413-422.
Zmeureanu R., "Cost-Effectiveness of Increasing Airtightness of Houses”, Journal of
Architectural Engineering, Vol. 6 (3), 2000, pp. 87-90.
45 OF 46
Zuercher C.H. and Feustel H., “Air Infiltration in High-Rise Buildings”, in Proceedings:
Air Infiltration Reduction in Existing Buildings, 4
Air Infiltration and Ventilation Centre Conference,
September 26-28, 1983, Paper 9.
46 OF 46
... The interdependency of airtightness, ventilation, and energy use has been demonstrated in studies on the interaction between these parameters (Sherman & Chan, 2004). As a result, the first step in reducing energy consumption and air quality issues associated with air infiltration is to understand buildings' airtightness. ...
... Those measures improve the thermal performance of buildings, but, on the other hand, they may result in air renewal decrease. In other words, if no ventilation system is implemented to provide the dilution of indoor contaminants, the IAQ may suffer (Sherman & Chan, 2004). The condensation of damp air in contact with cold envelope surfaces due to airtight envelopes and a lack of ventilation frequently cause mould. ...
... Its importance resides in the fact that it is the main building feature that impacts air infiltration (Sherman & Chan, 2004) and, thus, measuring and characterising airtightness is key to assessing the air infiltration through the building envelope. ...
Only after the oil crisis in the 1970s has there been an undeniable growth in the level of interest in reducing the use of fossil fuels. In this regard, the countries of the European Union predict a sustainable, competitive and decarbonised energy system by 2050. For that purpose, it is crucial to address the energy efficiency of buildings, which are responsible for a significant percentage of energy consumption, and, therefore, greenhouse gas emissions. The potential seems obvious, considering that, according to the European Commission, 75% of the buildings in Europe are energy inefficient. In this context, uncontrolled airflow through the building envelope, or air infiltration, is a phenomenon that involves energy loss since it needs to be conditioned at the interior comfort conditions. Air infiltration is also related to the ventilation of buildings and, thus, has an impact on indoor air quality. In this sense, a relationship is established between air infiltration, energy impact and ventilation. Since airtightness is the main characteristic of the envelope that determines air infiltration, its study is essential to understand the performance of the building envelope. However, this issue has not been addressed in depth in most countries with temperate climates, and, specifically, in Spain, given the traditional dependence of air renewal on air infiltration. Currently, the implementation of controlled ventilation systems guarantees adequate indoor air quality in residential buildings. Therefore, air infiltration is no longer necessary as an air renewal source and, consequently, its reduction is a priority to achieve nearly zero-energy buildings (nZEB). In this sense, the airtightness characterisation of the envelope of existing buildings is essential to prioritise efforts and determine strategies for the renovation of the existing building stock according to the decarbonisation objectives which have been set at a national and international level. To meet these needs, the general objective of this work focuses on the characterisation of the airtightness of the envelope of residential buildings in Spain, framed within the INFILES research project (BIA2015-64321-R), funded by the Ministry of Economy and Competitiveness of the Government of Spain. As a starting point, the state of the art was addressed and the airtightness regulatory frameworks in Europe and North America were compared. Additionally, the main databases and protocols for evaluating the airtightness of the envelope were assessed. Its analysis allowed the identification of the weaknesses, threats, strengths, and opportunities of these practices. Likewise, the main keys for the development of a common outline at the international level were detailed. The characterisation of the airtightness of the envelope of the residential building stock in Spain was carried out in a representative sample of cases that allowed the extrapolation of the information obtained. A quota sampling scheme defined by representative variables regarding the airtightness of the envelope such as the climate zone, location, year of construction, and building typology, was proposed. In short, 401 cases proportionally distributed in single-family and multi-family dwellings, built in different periods, and located in several cities, were analysed. This sample generated a database that collects information on the complete characterisation of both construction features and the airtightness of buildings. To facilitate data management, a specific application was developed, which enabled the sequential and standardised storage of information, whose operation could be replicable in other contexts. The airtightness of the case studies was evaluated by means of fan pressurisation tests, according to the international standard UNE-EN 13829. In addition, a specific methodology was developed to guarantee the uniformity and reliability of the information gathered. Airtightness results were wide-ranging, which indicates the uneven performance of the buildings tested. Nevertheless, the distribution of the average values was in line with previously reported findings both in Spain and in other European countries. Different trends were identified by climate zone: the envelope of the cases located in areas with Oceanic and Continental climates was more airtight than that of the dwellings located in areas with a Mediterranean climate with a temperate cold season. On the other hand, contrary to what might be expected, no clear trend of airtightness improvement was observed with regard to the year of construction. The different regulations regarding the energy efficiency of buildings did not lead to a significant improvement of the airtightness of the building envelope. Therefore, it seems clear that the existing residential stock in Spain remains far from the standards and practices that are already a reality in other European countries. However, the recent regulatory limitation of the global air permeability of the envelope opens a window to a change in this trend and represents an opportunity for real achievement of nZEB. The relationship between different building characteristics and the degree of airtightness was addressed through statistical analysis. Although the limitation of the sample did not allow the development of a detailed study, a relationship was identified between the airtightness and the type of windows, the rolling-shutter system used, and the degree of renovation of the envelope. In this sense, the main leakage paths were located using infrared thermography, which allowed the identification of typical deficiencies of residential buildings in Spain. Turbulent flows through cracks located around windows and rolling shutters concentrated the main leakages. The inadequate design of the constructive solutions, as well as the careless workmanship of the joints between different elements, were pointed out as the main causes that prevent airtight envelopes, and, consequently, they pose the main challenges to face in the future. Finally, the energy impact of air infiltration on the heating and cooling demand in the cases studied was addressed, applying a simplified model for estimating the average air change rate. The results demonstrate the high energy impact of air infiltration, which reaches values of up to 25% of the heating demand. This confirms the substantial potential for energy-saving if the air permeability of existing buildings was reduced. The characterisation of the airtightness of the envelope of residential buildings in Spain contributed to filling the knowledge gap identified. Only in this way will future design criteria and renovation strategies of existing energy-inefficient buildings be approached in a realistic way, in accordance with the established decarbonisation targets.
... Air tightness is the fundamental building property that impacts infiltration. Infiltration, or air leakage, is the movement of air through leaks, cracks, or other adventitious openings in the building envelope (Sherman and Chan 2004). Air leakage occurs at joints of the building fabric, around doors and windows, cracks in masonry walls etc., as well as where pipes and cables pass through the building (Hall 2008). ...
... However, in old buildings, infiltration is the primary source of outdoor air to control adequate indoor air quality (IAQ). Improving air tightness will need to be coupled with a ventilation system in order to provide sufficient airflow (Sherman and Chan 2004), and the thermohygrometric behaviour of the walls should be understood and taken into account when increasing the air tightness of the building to anticipate any possible side effect. Spray-applied foam is commonly used to block air leakage at holes and cracks; when used in small quantities, is reversible with little impact on the surfaces to which it is applied (ASHRAE Guildeline 34 2019). ...
Technical Report
Full-text available
This document provides a technical guideline for an energy audit of a historical building to support its energy and environmental improvement (as shown in the Energy Audit Process Flow schema based on the EN 16247-2:2014, see par 1.4), from the analyses to the design stage up to the Energy Performance Contracting implementation. Each section of the guideline can also be used as technical specification for tender activities, along with the related template Annex.
... In the summer, infiltration can bring humid outdoor air into the building, which is also caused by wind, stack effect, and mechanical equipment within the building. Poor airtightness can lead to large infiltration rates, eventually causing an increase in energy consumption [8]. According to the Clean Energy Business Council in 2018 [9], residential buildings consume 18% of the GCC's overall energy consumption and 43% of its electricity. ...
Full-text available
If infiltration is uncontrolled and admits unconditioned air, the results will be undesirable. Controlling this problem will increase thermal comfort and decrease energy consumption. The aim of this paper is to assess the performance of different materials used to improve airtightness, which will increase energy efficiency. This research primarily adopted an experimental approach. A typical residential building in UAE was chosen as a case study. Current airtightness status was measured using a blower door test and infrared technique. Six commonly used materials used for airtightness in UAE were identified and applied in different zones of the building envelope, including exterior walls, door and windows. The test was run before implementing airtightness strategies, following which they were applied for one year. Overall performance and energy reduction were monitored to identify how consumption fell by which method was the most efficient. The results indicate that energy was 3% when applying the 6 different airtightness strategies. The base case energy consumption was 64,287 kWh per year. The energy consumption then decreased after applying the sealants to 62,341 kWh per year. Future recommendations are made to enhance airtightness in residential buildings in a hot and arid climate.
... The type of flow through the ALPs is characterized by the exponent, n. In theory, the air flow exponent varies continuously from 0.5 for turbulent flow to 1.0 for laminar flow [20,57] . Therefore, the power law was adopted as the basic air infiltration model in this study. ...
Full-text available
Air infiltration through building envelopes has a considerable impact on the comprehensive performance of buildings, especially in terms of their energy demand and indoor air quality. Therefore, it is important to accurately predict building air infiltration rates under various scenarios. High airtightness is one of the typical characteristics of passive ultra-low energy buildings. With the rapid application of passive technology in building energy efficiency, the airtightness of new urban buildings has been significantly improved. The centralized air leakage path distribution assumption of current prediction model for building air infiltration rate is inconsistent with the actual situation of high airtightness buildings, which reduces its prediction accuracy and application range. Therefore, it is of great practical significance and academic value to carry out the research on the prediction model of air infiltration rate of buildings with high airtightness. This paper presents an air infiltration prediction model for single-zone buildings with adventitious openings. The building envelope was broken down into permeable parts and impermeable parts, and the air leakage pathways were assumed to be uniformly and continuously distributed in the permeable envelope. A linear pressure distribution over the building facade was assumed, and the airflow rate was integrated in the vertical and horizontal planes to theoretically predict the air infiltration rate. The feasibility of the proposed model was tested by comparing the air infiltration rates simulated by this model with those determined using the tracer gas attenuation method of an airtight building. The initial test results suggest that this model is mathematically robust and is capable of modeling the air infiltration of a building in a wide variety of scenarios. Reasonable agreement was found between the tested and simulated results. This study can provide basic theoretical support for the coupling performance analysis of high airtightness buildings.
... Hence, it is important to utilize relatively stronger winds at the higher levels for efficient natural ventilation [2]. To improve the performance of wind-driven ventilation, various devices such as wind walls [3], exhaust cowls [4], wind tower or wind catcher [5] were adopted in the previous studies. Meanwhile, a comprehensive review on wind-driven ventilation devices has been conducted in [6]. ...
Full-text available
This study aims to develop a new device, “cross-wind catcher”, for mid-to-high-rise affordable apartments for improving the ventilation performance in the tropics, where the natural ventilation is considered essential. We conducted a primary simulation study to find optimum wind catchers, followed by a field experiment using a full-scale experimental building to confirm the effects of the proposed device. The results showed that the proposed cross-wind catcher was able to increase the ACH by approximately 1.2 to 1.4 times compared with the control unit.
... For instance, the required levels of airtightness (≤0.6 h-1 @50 Pa) in Passivhaus dwellings help avoid condensation and conserve energy by reducing air infiltration. However, it is unclear whether an airtight building envelope has clear IAQ benefits [39,50] or not [51]. Nevertheless, occupants' satisfaction with IAQ and indoor humidity is better than those living in non-Passivhaus dwellings [44]. ...
Full-text available
Sustainable building design, such as the Passivhaus standard, seeks to minimise energy consumption, while improving indoor environmental comfort. Very few studies have studied the indoor air quality (IAQ) in Passivhaus homes outside of Europe. This paper presents the indoor particulate matter (PM2.5), carbon dioxide (CO2), and total volatile organic compounds (tVOC) measurements of the first residential Passivhaus in Latin America. It compares them to a standard home in Mexico City. Low-cost monitors were installed in the bedroom, living room, and kitchen spaces of both homes, to collect data at five-minute intervals for one year. The physical measurements from each home were also compared to the occupants' IAQ perceptions. The measurements demonstrated that the Passivhaus CO2 and tVOC annual average levels were 143.8 ppm and 81.47 μg/m 3 lower than the standard home. The PM2.5 in the Passivhaus was 11.13 μg/m 3 lower than the standard home and 5.75 μg/m 3 lower than outdoors. While the results presented here cannot be generalised, the results suggest that Passivhaus dwellings can provide better and healthier indoor air quality in Latin America. Further, large-scale studies should look at the indoor environmental conditions, energy performance, and dwelling design of Passivhaus dwellings in Latin America.
... Infiltration through the building envelope is a physical phenomenon which happens due to pressure difference between the interior and exterior environment. This action takes place through unintentional cracks in the envelope, and it depends on the wind velocity, temperature difference and crack position on the envelope [7]. Infiltration losses in modern residential buildings are near 0,1 1/h whereas the minimum overall air change rate (sum of infiltration air change rate and natural and/or mechanical air change rate) must be at least at 0,5 1/h because of hygienic minimum requirements. ...
Conference Paper
Full-text available
To achieve the goals set by the European Union (EU) in terms of energy savings and decrease of the greenhouse gas emissions, the member states of the EU obliged that they would increase the number of the Nearly Zero-Energy Buildings (NZEB). That increase has to be accomplished not only by building new NZEBs but also by deep energy renovations of old ones. One of the criteria to fulfil NZEB demands is to ensure the airtightness of the building, which is usually tested by an air pressurisation test. As there will be a great number of buildings which must undergo the testing in the near future, it is important to find a way to speed up the testing for the multizone buildings. This paper presents research based on the results of building zone airtightness estimations based on a fan pressurisation test by the so-called Blower Door system. It is a part of research whose goal is to develop a predictive model of the air-change-rate for a multizone building based on randomly tested zones. The paper describes a connection between the airtightness of tested building zones and elements which make up these zones. This research is crucial as a starting point for the development of the predictive model in order to define the parameters that affect these measurements. Statistical analysis was performed not only for the analysed results but also for an exploration of the connection between the zone elements and the zone's airtightness as well. The paper presents a hypothesis based on the abovementioned research and proposes further research for the future development of the model.
Full-text available
The novel Pulse technique measures the building airtightness in a dynamic approach, involving a low-pressure pressurisation process, typically in 1–10Pa. It is known that the wind effect is one of the main sources of uncertainty for airtightness testing. The literature review revealed that the wind impact on measuring building airtightness has been explored in relation to the steady pressurisation method (i.e., the blower door), while there is limited research investigating the validity of the Pulse measurement under various natural wind conditions. In this study, in total, 423 Pulse tests were performed to measure the building airtightness of a five-bedroom dwelling located at the University of Nottingham, UK, under natural wind conditions. The viability of the Pulse technique for delivering airtightness measurements under different wind conditions was assessed, including the impacts of the wind speed and the wind direction on the Pulse measurements. Based on the measured air leakage rates, the threshold of maximum wind speed that led the Pulse measurements to have repeatability greater than ±10% is 5.0 m/s at 2.2 m above ground level (i.e., approximately equivalent to the meteorological wind speed of 7.9 m/s). On the other hand, the directional wind study showed that at lower wind speeds, the wind direction has a lesser influence on the Pulse measurement than the wind speed itself. Practically, multiple Pulse tests are recommended for minimising the wind impact on building airtightness measurement when adverse wind conditions are present.
Energy demand for building heating, ventilation, and air conditioning account for a significant fraction of the global electricity demand. With global average temperatures projected to increase throughout the twenty-first century, building energy demand and consumption are also slated to increase. Even though the importance of climate effect on building energy management has been identified there has been few studies conducted to date to estimate the sensitivity of electricity demand for air conditioning to the climate variability, especially for the tropical weather conditions. The localized studies are of more importance in this regard, as the air conditioning load vary with the local ambient conditions. The impacts of climate change on building air conditioning energy demand for existing buildings can be reduced by establishing future energy demand patterns and using passive cooling strategies. Hence, this research aims to establish a relationship between the current energy demand patterns for air conditioning in a selected existing building and ambient temperature changes, thereby establish energy demand patterns. A suitable sample building was selected for the study and the indoor thermal comfort data, outdoor environment conditions and building energy consumption patterns are monitored and hourly data were collected. Climate conditions and the cooling load variability of the building were studied theoretically and the relation between climate conditions and energy consumption patterns were analysed. The indoor temperature and cooling load showed high sensitivity to the outdoor temperature with maximum of 25% cooling load increase for 1 °C increase in outdoor temperature. Also, it was predicted that the current cooling load of the building will increase by 40–55% in 2050. Further, the effects of the short-term meteorological variability on the cooling degree days are calculated and its impact on the energy demand was established for the selected building, which could be used for predicting future energy demand patterns with the help of different climate change models.
Full-text available
Blower doors are used to measure the airtightness and air leakage of building envelopes. As existing dwellings in the United States are ventilated primarily through leaks in the building shell (i.e., infiltration) rather than by whole-house mechanical ventilation systems, quantification of airtightness data is critical in order to answer the following kinds of questions: What is the construction quality of the building envelope? Where are the air leakage pathways? How tight is the building? Tens of thousands of unique fan pressurization measurements have been made of U.S. dwellings over the past decade, and the available data have been collected into an air leakage database. This report documents what is in that database and then uses the data to determine relevant leakage characteristics in the U.S. housing stock in terms of region, age, construction type, and quality.
Canada Mortgage and Housing Corporation (CMHC) conducted a series of attic research projects from 1988 to 1997. Initially, there were few field test data to substantiate how attics dealt with air and moisture transfer. The CMHC research developed a test protocol for attic airtightness and air change testing and then proceeded to field testing of a variety of attics in different climatic areas. An attic model, ATTIX, was referenced against test hut data and used to simulate attic performance across Canada. The latest research project compared the performance of nominally identical attics, one of each pair with full, code-required venting and one with all intentional holes sealed. Results show that ventilation plays a relatively small part in the control of attic moisture and temperature but, conversely, rarely provokes major moisture problems. This suggests that there is no significant advantage in changing current Canadian attic code requirements, except perhaps by allowing more flexibility in venting design.
A method has been developed to estimate the air leakage through high-use automatic doors. This air leakage is specified as a function of the rate of use of the door, the door geometry, and the pressure difference across the door. Two studies were carried out to obtain these results. One was a laboratory study of the discharge coefficients of doors of various geometries. The other was a field study of the times when automatic doors are open as a function of use. The results of the field study were analyzed and combined with the discharge coefficients that were measured in the laboratory study. The result was an air flow coefficient that is a function of the number of people using a door each hour. The designer can use this coefficient with the pressure difference across the door to estimate the rate of air leakage through the door.