Repeatability of the European Standardized Method for Measuring Sound Reflection and Sound Insulation of Noise Barriers

The EN 1793-5 and EN 1793-6 standards have been in use for many years as a method for measuring the intrinsic characteristics of noise barriers installed along highways and railways. They require a sound source and a grid of microphones, to be placed near the barrier and in free field conditions, according to predetermined distances. In principle, small errors in positioning the sound source and microphone grid may affect the results obtained. An international round-robin test was carried out in 2012 to evaluate the repeatability and reproducibility of the method, but until now no studies have been carried out to evaluate and compare the repeatability of laboratory versus in-place measurements performed with the same equipment and its variance when an imperfect positioning of sound source and microphones is taken into account. In the present work, multiple series of sound reflection index and sound insulation index measurements performed on noise barriers of the same kind installed in the laboratory or along a highway are presented. The measurements were repeated in different ways: (1) in the laboratory, leaving the source and microphones unmoved to assess the repeatability of the results and of the measurement system under controlled conditions; (2) in the laboratory, repositioning for each measurement the source and microphone grid to assess the robustness of the method under real conditions but in a controlled environment; (3) in situ, along a highway open to traffic, repositioning for each measurement the source and microphone grid to assess the repeatability of the method under real conditions in a critical environment. In both reflection index and sound insulation index measurements, the standard deviation on single-number ratings in all cases examined is well below the value presented in EN 1793-5 and EN 1793-6, which was obtained from statistical analysis of the international round-robin test performed a dozen years ago, suggesting that expert operators with state-of-the art equipment can achieve much better results now.