SummaryThis article presents a perceptual comparison of the loudness of a large range of 64 different typical "living room" sounds transmitted through two different walls: (1) a light-weight wall, composed of gypsum boards mounted on metal "C" studs and (2) a heavy-weight wall, built out of lime sand bricks plastered on one side. The two walls had different (laboratory measured) sound insulation spectra, but their Rw + C50−5000 rating was the same: 52 dB. Compared to the heavy-weight wall, the massspring-mass-type light-weight wall had better sound insulation properties in the middle frequency range and worse at low frequencies below 100 Hz and above 3150 Hz. Listening subjects had to choose the loudest stimuli between a pair of sounds (as transmitted through the lightweight and the masonry wall) presented through headphones in random order with one repetition, following a two-alternative forced choice (2AFC) procedure. Two sets of listening tests were conducted in this study, playing stimuli at realistic sound level and on artificially overall increased level. The listening experiments revealed significant differences in subjective assessment between the two types of acoustic insulation. The results also infer that the discussed single number rating does not adequately correspond with people's perception.
The sound insulation spectrum is analysed of 18 double glazing arrangements facades, of which 9 double skin facades were measured in situ and 9 in a laboratory setting. The influence of the cavity thickness, the parallelism of the two glass panels, the absorptivity of the cavity and the effect of the size of ventilation slots are investigated. The results are compared with double layer wall insulation prediction models. Also a new, simple model is proposed that predicts the sound insulation of naturally ventilated double skin facades, based on the coincidence frequency, the structural resonance frequencies, the cavity resonance frequencies, the façade construction, the dimensions the and material properties. The model predictions are validated by measurement data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.