The University of Gävle is currently involved in a project on saving energy in historic buildings (churches). An important factor in the determination of the natural ventilation rate is the adventitious leakage of the envelope. Measurement of leakage is therefore a key feature of the investigations. It was decided to adopt a new technique developed at the University of Nottingham (UNott). It is a pulse technique compared to the conventional steady technique.The conventional technique consists of generating a steady and high pressure difference (50 Pa) across the envelope by means of a fan. Such pressures are rarely encountered in ventilation and this leads to errors in the low-pressure leakage. Furthermore the use of the conventional blower door technique in churches is difficult due to their large volume and the need to replace the doors.The underlying principle of the UNott technique is described and examples of results are given. The most important advantage of the Unott technique is that the leakage is determined at the low pressure differences that are encountered with ventilation e.g. 4 Pa. This is made possible primarily by the fact that the effects of wind and buoyancy at the time of the test are eliminated by taking account of the pressure variation before and after the pulse.For measurements in large buildings, a number of identical piston/cylinder units have to be operated simultaneously. The University of Gavle has developed a system whereby up to seven units can be used. Such a number is required for a leaky church and this is the first time this has been done.