CEN/TC126/WG1 is dealing with the development of the European standards for measuring the sound insulation of building elements that are required for the implementation of the Construction Products Directive. Its activity is mainly related to the revision of the ISO/140 series of standards and it works in close co-operation with ISO/TC43/SC2/WG18. One of the standards being revised, ISO 140-3 (1995), describes a laboratory method for measuring the airborne sound insulation of building elements such as walls, floors, doors, windows and facade elements. Compared with the previous version, the new standard gives more detailed specifications for the installation of the test objects and for the generation of the sound field in the source room. It also extends the frequency range of the measurements both at low and high frequencies (50-5000 Hz). There is a need, however, to assess the uncertainty in the acoustical measurements performed according to the standard due to random and systematic influences.
The aim of the project was to determine the repeatability and the reproducibility of laboratory measurements of the airborne sound insulation of partitions used in buildings carried out in accordance with the test method described in the draft international standard ISO/DIS140-3 (1995).
The repeatability values, r, for both specimens, were much lower than the tentative values given in Annex A of ISO/DIS140-3. The reproducibility values, R, were very different for the two specimens. For the double wall, they decreased with frequency in the low frequency bands and increased with frequency, up to 12 dB, in the middle and high frequency bands above 315 Hz. At the high frequencies, the reproducibility values for the double wall were much larger than the tentative values in Annex A of ISO/DIS140-3. For the single wall, the reproducibility values decreased with frequency and were very similar to the tentative values given in Annex A of ISO/DIS140-3.
The results of the intercomparison have been transferred to the relevant standardisation bodies and it is proposed to update ISO/DIS140-3 with the repeatability values obtained. Further investigations on the causes for the poor reproducibility of the measurements on the double wall are recommended, as well as new measurements with different specimens.
A total of 24 laboratories from 11 different European countries participated in the intercomparison. Each laboratory made measurements on two types of wall: a single wall one with medium sound insulation, and a double light-weight wall with high sound insulation. Both walls were built with gypsum plaster boards on a steel frame supported with studs. Guidelines for the construction of the walls and for carrying out the measurements were prepared by the co-ordinator of the project and sent to all the participating laboratories. In order to ensure that the walls were as identical as possible in each laboratory, all the materials for the walls came from one single stock of production and the same person supervised the construction of the walls in all the laboratories.