The stationary experiment using a modal impact hammer is applied to measure
the driving and cross-point FRF in m/N s2 unit at the powertrain mounts, and noise
transfer functions (NTF) in Pa/N unit from each mount to the vehicle interior. This experiment is performed identically on the two vehicles with opposing rumble quality. By examining these response functions, it may be possible to determine the signi®cance of structure-borne path contribution. Fig. 12 shows the comparison of
typical FRF on the body and engine side of the transmission mount in the noisy and
quiet vehicles. Up to 1600 Hz, no substantial dierence can be seen in the response amplitudes. This implies that the opposing rumble quality of the two vehicles is not
due to the variance in the mounting property. In addition, Fig. 13 compares the
NTF from the timing belt and transmission mounts to the ¯oor microphone position.
Within the expected variability of the measurements for this type of complex structure,
the two sets of response spectra are considered comparable. Hence, we conclude that
the structure-borne paths do not play a signi®cant role in dierentiating the quality of rumble in these two vehicles. However, it is still believed that the structure-borne
path is the primary mechanism for transmitting rumble from the engine to the interior
as suggested in the spectrogram results and also shown in the subsequent analysis. In
fact, it is also noted that the FRF and NTF functions exhibit sensitivities in certain problem frequency ranges given in Table 1, such as 490±600 and 710±770 Hz. The
existence of these sensitive regimes may provide opportunities to reduce the transmission of rumble by structure-borne path control when source treatments are not
viable.