The well-known Kundt's tube and reverberant room method are often used for measurement of acoustic absorption properties of samples under laboratory conditions. Several in situ measurement methods exist, but most of them are limited in frequency range, require large samples and/or are vulnerable to background noise or reflections. The PU in situ impedance method [1, 2] has been used successfully on relatively small samples (> 0.1 m 2) in a broad frequency range (300 Hz − 10 kHz) under reverberant conditions (e.g. a car interior or a concert hall), see e.g. [3, 4, 5, 6, 7]. The small source-sample and probe-sample distance are the main reasons for the relative small sample size requirement and the low influence to background noise and reflections. However, in some cases the procedure shows artefacts because all the reflection at the top of the sample is considered, not taking into account wave propagation in the material. In this research the principle of measuring intensity instead of impedance is investigated. To eliminate near field effects an extrapolation technique is introduced that combines several measurements. The result is a technique to measure the absorption coefficient without knowledge of the material. The methods are examined theoretically and verified with experiments.

In this paper the measurement, in situ or under free-field conditions, of the surface impedance and absorption coefficient is investigated. Numerical simulations of the measurement of impedance above al ocally reactive surface is performed with the Boundary Element Method (BEM). Experiments are also made in asemi-anechoic chamber and in ar egular office room. Three calculation methods used to obtain the surface impedance of an absorbent surface are described and compared, twoo ft hem being iterative.T he first, referred to herein as the "q-term", relies on an exact description of the sound field above an infinite locally-reactive plane. The second, the "F-term", relies on an approximation for large values of the argument k | r 2 | in its equations. The third, the "Plane Wave Approximation (PWA )",isasimplification of the spherical wave reflection which considers that the reflected wave has its amplitude and phase changed by asimple planar reflection coefficient. The "F-term" and the "Plane Wave Approximation" methods also assume an infinite locally-reactive plane. The three calculation methods are compared, the differences in the found results are discussed. The three methods are compared mainly for small sound-source to sensor distances (| r 2 | = 0.3m)and it is seen that theytend to convergeasthis distance increase. This comparison is relevant to in situ impedance measurements, since abad choice of the calculation method may lead to apoor result. PACS no. 43.58.Bh, 43.20.El, 43.55.Ev Received2May 2011, accepted 18 September 2011.the sound pressure at twol ocations close to the surface of the absorber.T he particle velocity wasc alculated by the gradient of the sound pressure. The authors assumed the acoustic field to be formed by plane waves. The in situ measurements were compared with impedance tube results. Substantial deviation wasf ound below5 00 Hz, which can be attributed to an insufficient spacing between the microphones and to the assumption of plane waves. Li and Hodgson [2] also used twom icrophones, assuming that the reflected spherical wave fronts have its amplitude and phase changed by aplanar reflection coefficient. Despite the assumption regarding the reflection, asurface impedance closer to that obtained with the phenomenological model wasestimated. The results in reference [2] were compared with those in reference [1]. The assumption in reference [2] is the same as that described in section 3.3.Ac ommon weakness of these methods is the lack of a good description of the sound field above the surface. The ©S.Hirzel Verlag · EAA 1025 ACTA ACUSTICA UNITED WITH ACUSTICA Brandão et al.:M easurement of free field surface impedance Vol. 97 (2011)

Sound source localization techniques in a car interior are hampered by the fact that the cavity usually is governed by a high number of (in)coherent sources and reflections. The results demonstrate that even with a single sound source, the 3D intensity streamlines are strongly bending, suggesting that far field techniques do not point towards the sound source.

With PU probes the sound pressure and acoustic particle velocity can be measured directly. Over recent years, the in situ surface impedance method, making use of such a probe, has proven to be an alternative to Kundt's tube measurements for product development type of work. The in situ method can also be used for the purpose of quality control on the acoustic material, be it during manufacturing or assembly, ensuring the best possible way to monitor the practical effectiveness of the acoustic package designed earlier on. In order to assess the variance of the acoustic package material leaving the assembly line, a relevant number of samples were taken over time. The quality of both the headliners, and the passenger seats were measured, of 25 cars of the same type. The robustness of the measurement method will be discussed, and the results will be presented.

A novel measurement tool is developed that is capable to measure the reflection coefficient of acoustic materials in situ and thus in real live situations such as a car. The measurement tool is a combination of two novel methods, the surface impedance method [1], [2], and the mirror source method [3]. The surface impedance method measures the acoustic impedance close to the surface of an acoustic absorbing material. The method is very sensitive for highly reflective surfaces [1], [2], [6]. The mirror source method uses a miniature monopole sound source that is placed close by the acoustic reflecting material. A particle velocity microphone (a Microflown [4], [5]) is placed close to the monopole source in such way that its sensitive direction is aiming at the acoustic reflecting material and its non sensitive direction is aiming at the source. This way it is only measuring the 'mirror source': the reflected image of the monopole sound source. From the ratio with a reference measurement on a fully reflecting plate the reflection coefficient can be determined. In this paper some results of both methods are shown and it also shown that there is a minor effect of reflecting surfaces in the environment on the measured absorption curves. This observation paves the way for application of the described method in cars.