Carletti E., Pedrielli F.

In recent years the noise labelling has become mandatory for most machines and equipment and the knowledge of the uncertainties accompanying the sound power results has become essential either when the conformity with a specification or a limit value must be tested, or when values have to be compared to each others. In the existing sound power standards, the aspect of uncertainty is presently handled in different ways but a review is in progress, aiming at harmonising all the current methods according to the GUM document. Referring to the three ISO procedures based on sound intensity, it seems to be a very hard task to evaluate the measurement uncertainties according to the GUM document primarily because measurements take place under widely different conditions. This paper aims at investigating more in detail the influence of some specific sources of error on the repeatability of the sound power results when the procedures according to ISO 9614 part 2 and 3 are applied. In particular, the effect of the source operating conditions and that of the sound intensity spatial sampling over the measurement surface are investigated. Measurements were carried out on different types of gear pumps mounted on a specific hydraulic circuit, under controlled conditions of installation and operation. On the parallelepiped measurement surface three different scanning paths were selected with a growing scan line density. After an initial warm-up period, the measurement was repeated 10 times in order to check the statistical repeatability of the acquisition. For each data set, the variation of the sound power levels over the time has been evaluated and the fulfilment of the field indicators has been verified.