A round robin test on the close-proximity method: Comparison of results from four different CPX trailers measuring tyre/road noise properties of 10 Swedish road surfaces

Abstract

The standardization of the CPX method and the reference tyres in 2017 has made it possible to measure tyre/road noise in a standard and repeatable way. However, the reproducibility of the CPX method is still not well known. A trailer comparison in the Netherlands in 2017 indicated that different trailers led to comparable results, but only two road surfaces were used, and all measurements were made close in time and location. In practical measurements, more variations in CPX equipment occur, and it is not possible to predetermine a specific measurement day. This study used four trailers of different designs, and measurements were made over a time period of 90 days, which may happen in practical measurements. The trailers and road surfaces are described, and variations between measurements are analyzed. The maximum CPX level difference between trailers when taking the sample means (average without corrections for different number of observations on the surfaces) was 1.57 dB for tyre H1 and 1.29 dB for tyre P1, which is comparable to the uncertainty values in ISO 11819-2:2017. A statistical analysis showed that the estimated main effects, adjusted for imbalance in the sampling design, for three of the four trailers did not differ more than 0.41 dB, while the fourth trailer, having an unreliable calibration, differed by 0.94 dB from the others. This article shows the relevance of the corrections found in the standards and that having an enclosure over the tested tyres leads to a slightly sharper peak in the frequency spectra.