Investigation of the in-situ measurement method for determining the acoustic reflection properties of surfaces

The dissertation is blocked till 24. june 2012

This doctoral thesis discusses the measurement of the sound absorption of materials with the standardized measurement technique according to the prestandard ÖNorm CEN/TS 1793-5 (Adrienne-method) in order to determine the acoustic properties (sound reflection and sound insulation) of noise barriers in situ. The specified measurement technique is based on the assumption of the sound-field of an ideal point source, disregarding the real properties of the loudspeaker. In this thesis measurements are inquired and geometrical and wave field theoretical investigations are made on several influencing factors such as propagation loss, directivity, group delay and position of the acoustic center of the loudspeaker. Investigations on uneven specimen at different measurement positions show, that already small structural depths cause significant standard deviations of the determined sound reflection coefficient. In case of occurrence of all these influencing factors in a single measurement, the deviation of the single-value-specification of the sound reflection DLRI is 0,9 to +1,1 dB. An improvement is achieved by using a smaller loudspeaker, thereby the described influencing factors of the loudspeaker are reduced. The required sound level causes a higher loading of the smaller loudspeaker, for which reason the distortion increases. By using the logarithmic swept-sine-technique these harmonic distortions can be eliminated from the impulse response. Further improvement is reached by using the adaptive suppression of the undesired sound; thereby achieving an increasing of the signal-to-noise ratio. In order to have more significant information of the farfield-properties of especially structured noise barriers, an increase of the distance between the measurement setup and the specimen within the valid range is necessary. This leads to a shorter undisturbed time-window for analyzing the impulse response, implicating the disadvantage of increasing the lower limiting frequency. The undesired reflections can be reduced or eliminated by using the Beamformer-technique or cepstral techniques. Hence the measurement range can be extended to lower frequencies. In addition to these improvements rules for the number of measuring points and the distance of the measurement setup in dependency on the structural depth of the specimen are recommended. Besides, the use of a particular In-situ classification permits a better attribution of noise barriers which have been tested In-situ.