Reproducibility and repeatability of measuring noise level reduction using an artificial noise source

Abstract

Buildings that are subjected to aviation noise require extra sound isolation measures in order to keep the indoor noise levels below a certain threshold. The difference in sound pressure from outside to inside a building is often quantified as a single number known as the noise level reduction (NLR). Generally, the procedures described in ASTM E966-10 are followed in determining this number for a façade, which is then used to determine the modifications required. The standard allows testing with one of two noise sources: an actual traffic source or a loudspeaker. An investigation is underway to statistically evaluate the repeatability and reproducibility of measurements taken on a façade for a “test house” with a loud speaker. The test house is a single-room structure that was designed and constructed with materials and methods of a residence in a mixed-humid climate. The repeatability is defined as the ability for a specific test to be implemented multiple times with comparable results, while the reproducibility is the ability for various test configurations allowed within the standard to yield comparable results. The results of this analysis can be used to improve testing parameters, further validate the procedure, and provide an estimate of confidence bounds.