In-situ measurement methods for characterisation and diagnosis of airborne sound transmission through multi-layered building partitions

The thesis concerns with the development of novel measurement methods for characterisation and diagnosis of airborne sound transmission through building partitions. Using standard tests, the airborne sound insulation of partitions can be measured as the Sound Reduction Index ‘SRI’. While the SRI provides the frequency dependence of sound insulation (or transmission), the local/spatial sound transmission through various paths in the partition is not known. If the contributions of different paths in the partition can be measured then any weak paths of sound insulation can be diagnosed. This would be especially useful in the case of multi-layered partitions where the sound insulation depends on the sound transmission through point connections/ribs/studs/frame, etc. present in the structure. While different theoretical models are in place to predict the sound insulation in presence of such elements, the experimental diagnosis of their sound transfer contributions remains fairly unexplored. Similar diagnosis problems are encountered by automotive industry while dealing with structure borne sources in the vehicle. In practice, Transfer Path Analysis (TPA) methods are extensively used in such cases for diagnosing the contributions of different structure borne sources at vehicle interiors. Application of such TPA methods for diagnosing airborne sound transmission is challenging on various counts. Firstly, the airborne source applies a continuous excitation on the receiver as opposed to structure borne sources which are typically discrete. Secondly, for our study, the path contributions are desired which is difficult than measuring source characterisations. To address these issues, a novel TPA application Inverse-Airborne Source Contribution Analysis (I-ASCA) is devised which employs a patch based discretisation of the source receiver interface for the diagnosis of airborne sound transmission through partitions. Using such discretisation, the airborne excitation on the partition can be inversely characterised by blocked forces and the source contributions can be measured. Additionally, a new methodology Inverse Path Contribution Analysis (I-PCA) is outlined which allows for measurement of path contributions. These methodologies applied to the case of single and double layer partitions excited by airborne source and the accuracy of the methods was found to be within 2-3 dB of the measured response in general up to a maximum of 1 kHz under the tested grid size. The accuracy of the method is thus strictly linked to the discretisation size. A sampling criterion of λ_b/2 was found to be sufficient which is less demanding than sampling criterion utilised by finite element methods. The methods can be applied to >1 kHz range if the discretisation can be made finer. To improve the practical application of the methods, the Direct-Airborne Source Contribution Analysis (D-ASCA) is presented which allows for direct characterisation of the airborne excitation using contact pressures. The method is much faster than I-ASCA in providing source contributions however path contributions cannot be measured using this method and the accuracy of the method is also found to be within 2-3 dB. D-ASCA application has been presented for the case of commercial single and double casement windows. Using careful assumptions, it is possible to estimate the path contributions of the glazing and frame in the windows from the source contributions. The diagnosis allows comparing the path contributions in frequency regions up to 1 kHz and the weak path is identified. This shows the potential of the method in identifying the weak elements of sound insulation which can be used as a complement to the SRI data and can provide cues for improving the sound insulation of the partition. Lastly, an in-situ measurement method for airborne sound insulation measurement is presented which can be applied when a pressure doubling occurs at the surface. A novel approach has been presented to assess whether pressure doubling occurs on the surface and calculating the blocked pressures. This allows one to measure SRI in-situ using diagnostic measurements without the need of a separate standard test for measuring SRI. This showcases the versatility of the approaches in that the frequency dependence (SRI) and spatial dependence (path contributions) of sound insulation/transmission can be measured within a single approach.