There are three essential steps to achieving airtightness:
design for airtightness
build for airtightness
test for airtightness.

With complex designs, or when the project team is unfamiliar with designing and constructing tight buildings, it is advisable to seek advice from an expert at each of these three stages.

The test procedure
Air tightness tests should be carried out in accordance with CIBSE Technical Memorandum 23, and are usually carried out by an outside consultant appointed by the main contractor. The method used is 'pressure testing'.

Pressure-testing uses a large fan system to pressurise the building. The system is connected via flexible ducting to a wooden template that is temporarily sealed into an existing doorway. All external doors, windows and trickle ventilators are closed and all the internal doors are wedged open. Mechanical ventilation openings are sealed with polythene sheet and tape or other appropriate materials. Smoke extract fans/vents are closed but not sealed, and other integral openings (eg lift shafts) are left unsealed. Checks are made during the test to spot and correct any extraneous effects, such as a window or door blowing open, or any temporary sealing failing.

A test can take as little as two hours for a straightforward case, and good preparation by the contractor makes all the difference. More complicated or less well-prepared cases can take up to a full day.

Air leakage audit
This procedure can be carried out prior to airtightness testing. It will highlight any potential air leakage paths and enable remedial work to be carried out whilst access is still relatively easy. Air leakage audits are also undertaken when a building fails the test and again, is used to point out problem areas. BRE uses three audit tools and selects the most appropriate combination for the job in hand.
Small smoke tubes locate specific air leakage paths from inside the building. Identification of air leakage is easier if the building is under depressurisation using a fan pressurisation system operating in reverse, or using the building's own extract fan.
Infrared thermography detects the locally cooled areas where external air enters the building. This method is particularly useful for detecting air leakage paths quickly in high-roofed areas. However, it is only effective when the external temperature is several degrees below the internal temperature. Again, this is best carried out with the building under depressurisation.
Large-scale smoke tests can be used in some cases. This involves filling the building with smoke, then using a fan pressurisation system or the building's own supply fan to pressurise the building. The smoke is forced out through the gaps in the building envelope and observed from the outside. The results can be recorded on video camera.