Erratum

Abstract

Erratum: “Separation of Resonant and Non-Resonant Components-Part I: Sound Reduction Index” Jeffrey Mahn There was an error in the calculation of the shape factor for the non-resonant radiation efficiency σNR according to Annex B of EN12354 which was used for the evaluation presented in the paper. Although there are other formulas for calculating the radiation efficiency, it seemed reasonable to assume that a person calculating the apparent sound reduction index using EN12354 would use on the equations presented in the standard. The error in the calculation of σNR affected the calculations of the resonant component of the sound reduction for methods including: Method Gerretsen (eqn 2), C proposed (eqn 10), C A (eqn 16), C AnnexB (eqn 14) and C Villot (eqn 17). The correction of the error resulted in the resonant component of the sound reduction index calculated according to Method Gerretsen to be predicted in 5 of the 18 1/3 octave bands evaluated for the steel panel instead of none of the 1/3 octave bands. The resonant component for the MDF panel using Method Gerretsen could only be calculated in 2 of the 18 1/3 octave bands. When the measured sound reduction index was shifted to the bottom of the confidence interval, the use of Method Gerretsen predicted the resonant component in all of the 1/3 octave bands for the steel plate, but none of the 1/3 octave bands for the MDF plate which was evaluated. Therefore, the original conclusion that subtracting the non-resonant component from the total, measured sound reduction index is an unreliable means of calculating the resonant component was not affected by the correction of the error. Figure 8 of the paper compared the resonant components of the sound reduction index calculated using correction factors. The discussion of Figure 8 showed noted that the predictions of the resonant component of the sound reduction index were within 11 dB of each other with the exception of CAnnex B and CVillot. When the error in the calculation of the non-resonant radiation efficiency was corrected, the difference between the predictions was reduced to 8 dB as shown in Figure 1. [Figure: see text] Despite the error in the calculation of the non-resonant radiation efficiency, the overall conclusions of the paper remain the same. The application of the correction factors to the total, measured sound reduction index can result in a range of values for the resonant component of the sound reduction index as shown in the figure. Definitive guidance is needed for the determination of the resonant component of the sound reduction index in future revisions of EN12354 or ISO15712.