220 T. W. Jerome et al. The assembled test structure is suspended near a boxed 12-in speaker driver. This driver provides good energy output with low harmonic distortion at frequencies below 500 Hz, which covers the first four structural vibration modes. A quick frequency sweep for any given torque level reveals the resonance frequencies of the system in the current configuration. With that information, single-frequency resonance excitation or sweeps of finer frequency resolution and longer time intervals are performed to investigate each mode individually. 23.4 Results 23.4.1 Damping Acoustic excitation was used to induce free decay conditions for the first two resonances of the system with a fastened joint. Conditions for this test included torquing the fasteners to 23 lbf ft and exciting the system for about 10 s. Accelerometer signal acquisition began at least one second before the acoustic signal was terminated, and continued until the accelerometer signal approached the noise floor. A sample of this decay is shown in Fig. 23.15. This signal was trimmed and processed using previously developed methods to extract time-dependent damping [3]. A clip of the time series for the first two modal excitations are shown in Figs. 23.16 and 23.17. The blue curve is the signal from an accelerometer mounted on the flange of the long plate, and the red curve is the signal from an accelerometer mounted near the far corner of the same long plate. Both accelerometers are on the same half of the structure as split along its long axis, which is the bottom half of the structure in Fig. 23.8. Figure 23.16 shows the two accelerometers in phase, where the outside edge and the inside of the flange of the long plate are both either bending inward together or away from each other (see Fig. 23.4). In Fig. 23.17, the twisting motion of one corner is 180ı out of phase with the flange corner on the same long side of the structure (see Fig. 23.5). Damping curves for the first bending mode of this system in its configuration with very high torque are relatively constant in time. In Fig. 23.18, loss factor for the first bending mode is about 3:4 10 3. For the first torsional mode in Fig. 23.19, loss factor is about 1:3 10 3. These data show loss factors that are at or above the values obtained from initial boundary condition comparison data in Fig. 23.14. Loss factors in this section are greater than or equal to the corresponding modes of those in Fig. 23.14. For the fastened system, significant additional damping of some of the modes is expected due to friction losses at the interface. 1 2 3 4 5 6 7 Time [s] -3 -2 -1 0 1 2 3 Acceleration [ m/s2] Accel: L Flange Fig. 23.15 Decay time series after exciting the first bending mode
RkJQdWJsaXNoZXIy MTMzNzEzMQ==