Excitation and Measurement of BSR in Vehicle Seats

Buzz, Squeak, and Rattle (BSR) tests are commonly employed in the automotive industry as a diagnostic. The resulting signals are typically analyzed using established (historical) metrics such as A-weighted SPL or stationary loudness (ISO532B). However, due to the non-stationary nature of these signals, traditional metrics often fail to fully describe the signals in question. Compounding the issue is the fact that some specifications state that the test specimen is to produce ‘no objectionable noises’ when subjected to representative excitation. This is a very vague and debatable statement that normally cannot be settled by subjective observations alone. Within this paper, sound quality and statistical metrics are employed for several BSR signals acquired during forced response testing of a seat. Results are presented for two different boundary conditions and alternative signal processing tools are presented. Signal processing of non-stationary signals for an unoccupied and biofidelic mannequin-loaded seat are presented and compared.