An empirical study in predicting heavy-weight floor impact sound levels in the 31.5 Hz octave band using driving point impedance

In studies of floor impact noise, increasing attention has been paid to the lower frequency range below 100 Hz. Recently, several studies have suggested the importance of including the 31.5 Hz octave band in their evaluations. In this band, the floor impact sound is audible but has not been studied sufficiently. The accuracy of two types of so-called ‘impedance’ methods is investigated for the purpose of predicting heavy-weight floor impact sound levels in the 31.5 Hz octave band. Although impedance methods are widely used, they are not usually applied to the 31.5 Hz octave band. One of the two models is the usual method, in which the sound field of the receiving room is assumed to be diffuse; and the other is a possible variation in which no acoustical mode is assumed in the receiving room. Floor impact sound levels were estimated from driving point impedance levels measured in existing multiple-dwelling buildings and were compared with the measured floor impact sound levels. The results indicate that a hybrid of the two models, with their use depending on the length of the longer side of the receiving room, has the advantage of minimizing the maximum error. However, the hybrid method is unable to improve the general correspondence between measured and calculated values. A simple statistical correction corresponding to the clamping condition of the slab edges is determined to improve the correspondence. A more precise evaluation of the vibration transmission characteristics within a slab would be necessary for further minimization.