Sound Transmission Loss Through a Lightweight Double-Leaf Panel
Mosharrof, Mohammad Sazzad
MetadataShow full metadata
The topic of my PhD research is the sound transmission loss (TL) of lightweight double-leaf panels based on an analytical model. Computed values of TL are used to analyze the effects of various dimensions, parameters and boundary conditions of the panel. A double-leaf panel is usually made of two plates attached by a number of beams. Analyses are carried out on finite sized panels, where multiple separate cavities between the plates and the beams are taken into account. These panels are widely used in building, aerospace and shipbuilding industries. Although the construction and transportation of these lightweight panels are convenient, a drawback of such a panel is that it is easy to make the panel vibrate and the vibration propagates through the structure and radiates sound. There are many factors that influence the transmission of sound, which are the boundary conditions, material properties, dimensions of the plates and the beams. The effects of the variations of the panel parameters on TL through these panels is another topic of this thesis. The small scale (variation in apparently identical panel components) and the large scale variations in the panel parameters are taken into account. TL is computed for an airborne excitation. The Kirchhoff thin elastic plate equation, the Euler beam equation and the continuity conditions at each plate and cavity connection are used. Spring type connection is used for plate-beam-plate connections. The coupling operator K is found to be very crucial, which needs to be selected accurately through trial and error. The panel is subjected to simply supported, clamped or mixed boundary conditions. The Fourier series and the Galerkin methods are implemented. The boundary conditions mainly affect the lower frequency range and the effects are prominent for smaller panels. The first resonance frequency (ƒ₁) is the most significant parameter in the low frequency region. Splitting the cavities does not affect the ƒ₁ dominated low frequency region. Multiple partial cavity resonances occur when multiple cavities are considered. TL shows higher values above f₀ till a certain frequency ƒt. ƒt is found to be related to the cavity width. Materials with comparatively less Young’s moduli for the plates are recommended, and thicker and denser beams are also recommended. The cavities are recommended not to be too deep when no absorbing materials are used. A method of studying the effect of small scale variation in parameter values on TL variation is described, where TL is calculated for a range of values of parameters. ±5% variation in three parameters e.g. thickness of the radiating plate and beams, and the cavity depth, are used as an example. The joint effects of two parameters are also studied. The difference in the maximum and the minimum TL is used to quantify the effects. The analysis is done in 125 Hz, 250 Hz, 400 Hz and 800 Hz bands. The effectiveness of different parameters varies with the frequency bands. A regression model based on the Response Surface Method (RSM) is proposed to study the trend in TL variation for small scale (±5%) parameter variations. Effect of variations in seven parameters on TL variation is demonstrated as an example. The seven parameters are thickness of the two plates and the beams, the mass density of the plates and the beams, and the cavity depth. A method of optimizing TL is also described.