Structure borne sound in motor-vehicles using statistical energy analysis
Abstract
This Thesis is about the application of Statistical Energy Analysis (SEA) to the study of
noise and vibration in motor vehicles. Methods for predicting the transmission of
structural waves between body panels in motor cars are given. A complex joint technique
based on the wave dynamic stiffness approach is discussed in detail. This technique can
account for the complexities of the joint structures found in motor vehicles.
Transmission of vibration energy between panels connected by a curved section of panel
has been investigated using the complex joint modelling technique. Wave Intensity
Analysis(WIA) and SEA were used to calculate the coupled panel vibration levels based
on transmission coefficients calculated using the complex joint techniques. Energy Level
Differences (ELDs) were measured for comparison with predicted results using laboratory
models. SEA failed to accurately predict the ELDs except at low frequencies. However
WIA was able to predicted the ELDs with reasonable accuracy using an existing
expression for transmission at a panel corner joint.
Two SEA models of a small passenger car have also been constructed. One assumes
simpler joint configurations when calculating transmission coefficients and the other uses
the complex joint technique. Panel ELDs were measured using a small passenger car. The
SEA model using the complex joint method gives better agreement with measured ELDs
than did the model using the simpler joints. saloon Sound Pressure Level (SPL) as a result
of engine noise was measured. Both SEA models of the vehicle gave good agreement
between measured and predicted Saloon SPL. A noise problem associated with exhaust
misalignment is also investigated.