Unsupported browser

For a better experience please update your browser to its latest version.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

technical & practice

  • Comment

Envelope airtightness

In the first of a two-part analysis we examine the often forgotten benefits of improving airtightness in buildings

Architects frequently insist on betterthermal insulation, thermal mass, windows designed for better daylight and passive solar gain, shading systems and ventilation stacks. But envelope airtightness often seems to be forgotten, even in ostensibly 'sealed' buildings. Many clients assume their new building will be airtight, just as they are watertight. What kind of answers and assurances should they expect?

Does infiltration matter?

As buildings get better insulated, the proportion of heat loss attributable to ventilation increases; often accounting for more than 50 per cent of the design heat loss. Shortcomings can undermine low-energy strategies and lead to disproportionate reductions in performance, for example:

Energy wastage through unnecessary ventilation, heating and cooling loads

Problems with designed airpressurisation (for example of escape stairs), and in laboratories, medical facilities and computer rooms

Poor comfort conditions as systems struggle to deal with heat losses

Wasted energy due to increased temperature settings or the extended use of pumps, fans and ventilation, systems, etc

Routine oversizing of hvacsystems to reduce the risks, increasing their capital and running costs.

Trends towards prefabrication and subcontract packages have sometimes made things worse: with not enough effort put into the technical and management interfaces. In the worst case scenario, unexpectedly high infiltration rates can make a building unfit for its purpose.

Recent buildings

In the 1990s, pressure tests by bre and bsria revealed that British buildings could be very leaky. In the Probe 1 series of post-occupancy studies of recently-completed buildings1, 2, client comments on poor airtightness were widespread. In Probe 2, pressure tests were added; and only two out of the eight buildings met reasonable standards (see chart). Most of the air leakage was at eaves; at junctions between heavy and lightweight elements (see opposite page); through entrance and escape doors; around the frames of windows, rooflights and curtain walling (see page 51); and to a lesser extent through openable window elements. Holes where building services and structural elements pass through the envelope are also vulnerable. Motorised louvres for automated natural ventilation could also be very leaky (windows usually shut very much more tightly than building services dampers!); and in one building surveyed the occupiers had blocked them off. Reception areas were often draughty, requiring major alterations in four of the buildings.

How airtight is airtight?

British levels of airtightness are pitiable in comparison with good practice buildings in northern Europe and North America. After the 1973 oil crisis, research in North America and Scandinavia showed that air leakage could be decreased almost to zero:

Steel and timber-framed construction was leaky unless built with care. However, with major improvements in detailing and workmanship, the best examples - using techniques such as the polyethylene air-vapour barrier - showed that these, too, became extremely airtight

Heavyweight load-bearing buildings with in-situ finishes were more intrinsically tight, and the extra demands on workmanship and inspection were less onerous.

After development and field testing, the end result was that from 1980, 'best practice' envelope air-leakage standards in some countries were less than 0.15m3 per hour per m2 of envelope area at 50 pascal pressure difference for finished timber and concrete buildings, and less than 0.2m3 per hour per m2 for masonry. Compare this with 14 m3 per hour per m2 average for pressure tests of British buildings carried out by bre, as shown in figure 2.

Too much airtightness can be a worry, sometimes leading to airquality and moisture problems. Indeed, for a while 'sick building syndrome' was known as 'tight building syndrome', until tests showed that some of the suffering buildings were far from airtight. Most of the Probe buildings included engineered ventilation systems, so did not need infiltration to help them along, but they would have benefited from tighter envelopes. bre promotes the maxim 'Build tight, ventilate right'. The principle is that accidental air leakage should be avoided, while managed air flows should be designed in. Without mechanical ventilation systems, this may require trickle ventilation or a degree of designed-in infiltration to provide suitable background levels of air quality and humidity control.

Doing better

Achieving better airtightness is not simply about drafting performance specifications: it also requires designers and builders to understand how to achieve it in strategy, detailing, site quality control, and through the life of the building. For architects a good start is to draw a red line on all plans and sections of the building to indicate the surface at which airtightness will be maintained. This will not always be at the outer perimeter of the building, it will often run along the floor of a roof plant room for instance. To avoid moisture problems, it is also usually best to 'seal' near the inner surface of the envelope.

Then, at all points along these lines, one needs to determine how the airtightness will be maintained. This may require new interfacing components, for example effective wall cavity closers properly sealed to appropriate parts of windows and walls, and specially-formed gaiters to link the underside of profiled metal roofing to flat walling.

Designers should also check that the proposed details are buildable. This can often lead to changes to the detailing and sometimes the construction sequence. Often junctions as initially drawn cannot be made airtight in practice because, for example, a structural element is in the way.The requirements then need to be properly specified,implemented and monitored on site.


Testing is also desirable, especially while experience and confidence is being built up. There are currently some differences between the standards and methods of different testing organisations, so later this year cibse will be publishing guidelines (see page 493) which will also include indicative standards of today's good and best practice, which are likely to be as tabulated. Even though these do not reach the standards set by the best overseas practice, they represent achievable standards for uk practice today. They also make allowance for adequate background ventilation in naturally ventilated buildings.


Improvements in airtightness - through fabric insulation, windows and use of thermal capacity - are achievable and should be supported. Given that they are essential to good, all round performance, clients should be encouraged to make the money available, especially if it means that, if done properly, other things can be omitted. One example is the Elizabeth Fry Building, by John Miller & Partners at the University of East Anglia - which is examined as a case study in next week's article on airtightness. High insulation, well-specified windows, effective use of thermal mass and attention to detail on airtightness allowed a low-energy ventilation system to meet all the heating and cooling loads and perimeter heating to be omitted in all but a few corner locations. This contributed to the building's unusually high comfort and energy-efficiency, but at a normal budget4.

Bill Bordass is principal of William Bordass Associates


1 A series of post-occupancy studies undertaken under detr's Partners in Technology research and published in Building Services from July 1995 onwards. See also the Probe website, at www.usablebuildings.co.uk and www.building-focus.co.uk

2 W Bordass, R Bunn, R Cohen, P Ruyssevelt, M Standeven and A Leaman, The Probe Project: Technical Lessons from Probe 2, cibse National Conference 1999, Harrogate (4 October 1999)

3 cibse Technical Memorandum 23, Air Leakage Testing of Non-domestic Buildings, (in preparation).

4 M Standeven, R Cohen, W Bordass and A Leaman, 'Probe 14: Elizabeth Fry Building', Building Services 37-41 (April 1998)

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions.

Links may be included in your comments but HTML is not permitted.