Variability of low frequency laboratory measurements of acoustic isolation of floor/ceiling assemblies

It has long been recognized that building acoustics measurements performed in laboratories exhibit a high degree of variability between and within labs. Traditional metrics limited the use of data to above the 100Hz or 125Hz one-third octave bands. In recent years, focus has widened the range of frequencies of interest to the 50Hz one-third octave band or lower. Analyzing measurements of low frequencies in buildings is complicated by the physical reality of wavelengths that are similar in scale to common room dimensions resulting in modal behavior and the mathematical simplification of the wave equation in building acoustic calculations that assumes a diffuse field which discounts both phase and angle of incidence. Evaluation of common building isolation measurements of the same test floor/ceiling structure sample in the same lab and between labs across the frequency spectrum allows the comparison of the repeatability and reproducibility at low frequency ranges to those at high frequency ranges. Measurements and their associated metrics of various sources, including pink noise produced by loudspeakers, the standard tapping machine, and the impact ball are included in the analysis.