The air tightness of a structure has a direct impact on the natural ventilation rate and
will increase with wind speed and internal/external temperature differences. The
unnecessary air leakage of buildings in arbitrary locations leads to occupant discomfort
and higher energy consumption. The location of discomfort zones will vary with wind
direction and undefined air leakage routes. The magnitude of the discomfort will vary
with wind speed and outdoor temperature. No heating or mechanical ventilation system
can cope with high infiltration loads which change location with weather conditions.
Buildings should be designed and constructed to provide minimal air infiltration except
in locations which are part of the building design.
One of the key factors in providing energy efficient, comfortable buildings is
specification of an acceptable rate of air leakage and subsequent air tightness testing to
ensure compliance. This will help to ensure that clients are provided with buildings
which will meet their expected performance.
The air leakage of standard UK buildings is generally poor as revealed by BSRIA
Technical Note 7/92, "Ventilation Heat Loss in Factories and Warehouses" and BSRIA
Technical Note 8/95, "Air Leakage of Office Buildings." There is considerable scope for
improvement and this Specification presents the maximum air leakage rates that should
be specified for buildings and enclosures.
Different building types require different air leakage limits. For example, low energy or
air-conditioned buildings require a tighter specification than naturally ventilated
buildings. Archival stores and museums may require even tighter buildings for close
humidity control and prevention of ingress of external pollutants. Cold stores require a
particularly stringent specification to maintain food quality and to minimise energy
losses.
BSRIA’s Ventilation and Special Projects Section has undertaken whole building
pressurisation tests on 15 factory/warehouse buildings, 240 superstores, 44 large office
buildings, 4 archival storage facilities, 3 schools, 2 cold stores and a variety of
pressurised stairwells, builders ventilation shafts, floor voids and fire protected
compartments. The majority of the buildings were required to meet various air
tightness performance criteria of 1, 2, 5, 7.5 and 10 m3.hr-1.m-2 at a test pressure of 50
Pascals.
The air tightness specifications presented here are based on this extensive accumulation
of site test data and each specified level has been bettered in practice by a margin of at
least 40%. So there is clear evidence that these maximum air leakage targets can be
achieved - and be verified by testing.
In summary, BSRIA recommends the following for new buildings:
Type Maximum air leakage m3.hr-1.m-2 at 50 Pa
normal best practice
Offices
¨ naturally ventilated
¨ air conditioned/low energy
10
5
-
3
Factories/warehouses 10 -
Superstores 5 3
Museums and archival stores 2 1.4
Cold stores 1 0.5
Dwellings 10 5
Ó BSRIA Specification 10/98 Air Tightness Specifications 3
TERMINOLOGY
The rate of air infiltration into a building is often expressed as air changes per hour
with typical values of ¼, ½ or 1 air change per hour. The heating and cooling systems
in buildings are sized to cater for the loads imposed by air infiltration under design
external and internal conditions.
The design external conditions are usually taken at fairly extreme temperature values.
However, for assessing air infiltration rates it is also necessary to take relatively high
wind speeds to be sure that the systems will meet the loads under all but exceptional
conditions. The average air leakage of UK office buildings is 21.8 m3.hr-1.m-2, from
BSRIA Technical Note 8/95, "Air Leakage of Office Buildings". It is quite clear that a
design infiltration load based on 1 air change per hour would have been exceeded in
nine out of twelve typical buildings when the wind speeds were 12.6 m.s-l (strong
breeze) or above.
Air tightness testing involves the measurement of the airflow rate, Q50, required to
pressurise the enclosure to 50 Pascals. This pressure is low enough not to cause any
damage to the building and high enough to overcome moderate wind speeds. To relate
the measured airflow rate to an air tightness standard, the flow rate is normalised by the
envelope area of the building, S. This is defined as the area of walls and roof, wherever
the airtight surface has been established, which is usually at the inner surfaces of the
building fabric. This yields a value for Q50/S, the main parameter used in this
Specification. This normalised flow rate is also useful for assessing the suitability of
cladding, the air tightness of which is usually expressed per unit area.
The measured airflow rate can also be used to approximately estimate the area of gaps
in the building envelope through which the air would leak. This gives building
designers and contractors a better 'feel' for the required target. For example, a gap area
of 3 m2 is much easier to envisage than a Q50 value of 16 m3 s-1.
Air tightness standards are sometimes expressed as air changes per hour at test
pressures of 25 and 50 Pascals. For a single storey building of moderate size, the
following table compares such standards in terms of Q50/S values:
Test pressures Q50/S
m3.hr-1.m-2
1 air change at 50 Pascals 4.57
0.5 air changes at 50 Pascals 2.29
0.25 air changes at 50 Pascals 1.14
0.5 air changes at 25 Pascals 3.24
0.25 air changes at 25 Pascals 1.62
PERFORMANCE SPECIFICATIONS
OFFICE
BUILDINGS
Naturally Ventilated Offices
The air infiltration rate for a naturally ventilated office building should not greatly
exceed the rate of ventilation required for the occupants during high wind speed
conditions. During less than high wind speed conditions, the occupants should be able
to open ventilators or windows to provide adequate ventilation. The ventilation design
for naturally ventilated buildings is actually quite complex, especially during periods of
low wind speeds and moderate internal/external temperature differences. However, the
openings in the structure should be purpose made and mostly limited to occupant
control or fixed trickle ventilator openings in accordance with the selected ventilation
strategy. The air leakage of the structure should be at a sufficiently low level to avoid
causing draughts and discomfort, and increased space heating requirements.
4 Air Tightness Specifications Ó BSRIA Specification 10/98
A moderate air tightness specification of 10 m3.hr-1.m-2 is recommended for this
building type with all windows and trickle ventilators closed. During average wind
speeds the air change rate would be approximately 0.3 - 0.4 air changes per hour but
windows or ventilators could be opened, if required. The air change rate will rise
to > 1 air change during windy weather conditions.
If a high surface area to volume ratio is inherent in the design of a building, for
example by the inclusion of courtyards, then a tighter specification may well be
required. Buildings in exposed locations will require a more stringent specification.
Air-Conditioned and Low Energy Offices
Ventilation air required for the occupants and cooling to combat internal heat gains are
provided by air-conditioning systems and no natural air infiltration is required. Indeed
all natural air infiltration in such buildings is an additional energy load and can cause
local discomfort.
It is impractical to demand a perfectly sealed building but, based on test data, an air
tightness specification of 5 m3.hr-1.m-2 is clearly achievable and is recommended as the
maximum limit for air-conditioned and low energy buildings. For moderate to large
office buildings this would result in infiltration rates of approximately 0.15 - 0.2 air
changes per hour during average wind speeds and temperatures, rising to approximately
0.5 air changes per hour during high wind speeds.
FACTORIES/
WAREHOUSES
The average air leakage of UK factory/warehouse buildings exceeds 30 m3.hr-1.m-2,
which is excessive. At this value the heat loss resulting from the air leakage is likely to
be at least double the amount predicted using the data in the CIBSE Guide.
An air tightness specification of at the most 15 and preferably 10 m3.hr-1.m-2 is
suggested as a cost effective compromise. However, the purpose of the building should,
more appropriately, dictate the envelope integrity specification. For instance, some
warehouse buildings incorporate dehumidification for stock preservation and these
buildings would require a very much tighter specification to ensure that the required
internal conditions could be maintained. Similarly, air curtains designed to protect
building openings will not be effective unless the air leakage through the building
envelope is minimised.
SUPERSTORES These are air-conditioned buildings and should be built to an air tightness specification
of 5 m3.hr-1.m-2, or to comply with best practice, 3 m3.hr-1.m-2. The use of open
entrances for ease of customer access requires, among other design considerations, that
these structures strictly comply to this specification.
MUSEUM &
ARCHIVAL
STORAGE
BUILDINGS
Many of these buildings incorporate items which require very close control over
temperature and humidity, and the exclusion of pollutants. Where small-tolerance
control and high-grade air filtration and is required, an effective air tightness
performance specification will also be necessary. An air tightness specification of
2 m3.hr-1.m-2 is recommended and will be essential for such installations.
DWELLINGS There are a number of schemes which have been recommended for a number of
years. The Medallion 2000 scheme recommends an air tightness test requiring the
structure to meet a maximum air leakage of 7 air changes per hour when subject to
an internal/external pressure difference of 50 Pascals.
The above is quite a good standard, except that the standard of construction will need
to be better for small dwellings compared with larger dwellings. BSRIA would
therefore recommend an air tightness performance of 10 m3.hr-1.m-2, except for
mechanically ventilated dwellings which should achieve 5 m3.hr-1.m-2.
Ó BSRIA Specification 10/98 Air Tightness Specifications 5
COLD STORES The critical requirement to minimise air leakage into cold stores leads to a very
stringent criterion. An air tightness specification of 1.0 m3.hr-1.m-2 is provisionally
recommended, with the caveat that 0.5 m3.hr-1.m-2 would be preferred when this has
been demonstrated to be routinely practicable.
MISCELLANEOUS Ventilation / Builders Shafts
The air leakage of builders shafts as ventilation ductwork often runs into difficulties
with regard to specification and indeed achievement of a specification.
The HVCA document DW/142 entitled "Specification For Sheet Metal Ductwork"
recommends a maximum air leakage for low pressure Class A ductwork of
0.54 1.s-1.m-2 (1.94 m3.hr-1.m-2 ) at a pressure differential of 100 Pa and for medium
pressure Class B ductwork of 0.114 1.s-1.m-2 (0.41m3.hr-1.m-2 ) at a pressure differential
of 100 Pa.
The following table summarises these specifications for the same test pressures:
m3.h-1.m-2 at
100 Pa
m3.h-1.m-2 at
50 Pa
HVCA Class A ductwork 0.41 0.29
HVCA Class B ductwork 1.94 1.37
BSRIA very good building 3.54 2.5
BSRIA good building 7.07 5.0
Average UK office building 30.83 21.8
It would be unreasonable to expect a builders shaft to conform to medium pressure
ductwork and quite difficult to achieve low pressure ductwork standards. However, they
should not exceed the air leakage standard for a good building and preferably not
exceed the standard for a very good building.
Museum Display Cases
The current standard is based on a requirement to achieve 0.1 air changes per day and
is usually measured using tracer gas techniques. A pressurisation standard has not yet
been firmly established by BSRIA but may well be included in any revision to this
document. The reason for this is that the volume to surface area ratio changes rapidly
with size and the joints (potential air leakage paths) are clearly defined. The air
tightness standard is not constant to achieve 0.1 air changes over a wide size range and
the distribution/location of the air leakage paths is quite critical.
Floor Voids (Ventilation Plenums)
Where floor voids are used for ventilation plenums as used in displacement ventilation
systems, the BSRIA recommended air tightness criteria should remain as 1 litre per
second per square metre of floor area.
Pressurised Stairwells
The current British Standard should be used. This is BS 5588 : Part 4 : 1998, “Code of
Practice for Smoke Control using Pressure Differentials”.
6 Air Tightness Specifications Ó BSRIA Specification 10/98
DESCRIPTION OF TEST PROCEDURE
The air leakage characteristics of buildings are determined using an air pressurisation
technique. Air is supplied to the building over a range of air flow rates and at each the
resulting pressure differential across the building envelope is measured. This pressure
differential and measured air flow rate can be related by the equation:
Q = k . (Dp)n
where: Q is the air flow rate supplied to the building, m3.s-1
Dp is the pressure differential across the building, Pa
k is the air leakage coefficient, m3.s-1.Pa-n
n is an exponent normally between 0.5 and 1.0.
BSRIA developed this pressurisation technique to assess the air leakage of larger
buildings using a mobile test facility known as the "Fan Rover". This equipment
consists of a large fan unit mounted on a trailer, and driven using the rear power takeoff
of a Land Rover, thus avoiding the need for any intrusion into the electrical system
of the building under test. This facility is designed to supply air at flow rates up to
30 m3.s-l, and incorporates a flow grid that enables air flow rates to be measured down
to 3 m3.s-1 . This built-in flow grid consists of two tubes mounted across the unit
incorporating total pressure holes spaced at Log-Chebycheff intervals. The unit is
calibrated at BSRIA using standard anemometric and tracer gas techniques. A separate
“Wilson” flow grid is used to measure flow rates below 5 m3.s-1 when necessary. The
pressure differential across whichever flow grid is in use and across the building
envelope are both measured using Furness Controls Type FC014 Micromanometers
regularly calibrated by the manufacturers, augmented by regular calibration checks by
BSRIA.
Throughout the test periods, air temperatures are measured using PRT probes connected
to panel counters or a data logger. The accuracy of these probes is better than 0.2°C.
The internal temperature is averaged for the period of each test to provide a mean
internal air temperature. Similarly, the external temperature is averaged throughout the
period of each test.
Buildings are tested with all external doors and windows closed and with all internal
doors wedged open. Any natural and mechanical ventilation openings are also sealed
with polythene sheet and self-adhesive tape. Smoke extract fans or vents are left closed
but are not sealed and other integral openings such as lift shafts are left unsealed.
Ó BSRIA Specification 10/98 Air Tightness Specifications 7
DATA PROCESSING
The results of measurements directly associated with air leakage tests are initially
verified on-site. This consists of converting the pressure differences across the flow
grid into air flow rates using the calibration data. The pressure difference across the
building envelope versus the measured air flow rate is plotted on a log-log graph, and
the slope determined using a portable PC and printer. This immediate analysis provides
confirmation that the relationship between these parameters was generally as expected
and that nothing extraneous, such as a door or window left open, had occurred during
the tests.
The air flow test data is further processed to take account of two factors. First, an air
density correction is applied. This is determined from the air temperature and
barometric pressure at the flow grid. The second correction applied is a change in the
air volume flow rate that occurs if there is any difference in the temperature of the
supply air and that within the building. For example, during the pressurisation test,
outside air passes through the apparatus into the building and mixes with the inside air.
If the indoor air temperature is higher, the supply air expands so that the volume rate of
flow out of the building envelope is slightly greater than the measured air flow rate.
Following these corrections, a regression analysis is carried out on the pressure
differentials across the building envelope and the corrected air flow rate to calculate
values of ‘k’ and ‘n’ (see the equation on page 6). The correlation coefficient is also
calculated to indicate the 'closeness of the fit' of the data to the calculated relationship.
Using the calculated relationship, the air flow rate required to pressurise the building to
50 Pa is determined and then normalised with respect to the surface area (S) of the
building to yield values for Q50/S (m3.hr-1.m-2).
SMOKE TESTS
If the air leakage of a building under test is greater than specified, a smoke test can be
carried out to help identify the air leakage routes. The fluid from which the smoke is
generated is a food-grade polyglycol mixture often used in theatre and disco
applications. The building is pressurised by the “Fan Rover” while smoke is released in
all or part of the interior. Visual observations outside, as well as photographs and/or
video recordings, are made of the smoke egress from the building.
For large office buildings, it is usually more appropriate to undertake localised smoke
tests. This involves the use of a smoke generator with a ducted outlet which can be
directed at particular areas of the structure.