Modal Shaker Testing
With Modal Shaker Excitation testing the structure is excited at one fixed DOF, and Responses are measured at number of DOFs. FRF signals between response DOFs and excitation DOF are computed. With single input (excitation) multiple output (response), one column of FRF signals are measured. When the input channel count of the Dynamic Signal Analyzer is not high enough to cover all the measurement DOFs in one shot, the measurement sensors can be roved and the measurement can be repeated to finish all the required response DOFs.
To drive the modal shaker, an output channel from the Dynamic Signal Analyzer will be used to drive the amplifier of the modal shaker. There are many waveform types are available to drive the structure under test. Commonly, the Pure Random (Gaussian Random, or White Noise) can be the choice to drive the structure under test. Due to the nature of leakage, window, i.e., Hann window needs to be applied with this type of excitation waveform.
Another commonly used excitation waveform for Modal Shake testing is the Burst Random. The signal type is still random nature, but it will output the random with user defined percentage, and then keeps no drive sending out. This way, the structural response will decay within the zero-output duration of each block of data. In case the response does not decay enough, the percent of the burst random can be tuned to increase the zero-output period. With this achieved, there will be no leakage and thus no windowing would be required. And this is the main reason for the choice of Burst Random type of excitation.
Periodic Random and Pseudo Random are another category of random waveforms. The Pseudo Random is defined as an ergodic, stationary random signal consisting of energy content only at integer multiples of the FFT frequency lines (Δf). The linear spectrum of this signal is shaped to have a constant amplitude, but with random phase. The following figure illustrates this constant amplitude characteristics of the pseudo random signal.
Pseudo Random Signal Spectrum
When sufficient delay time is allowed in the measurement procedure, any transient response to the initiation of the signal will decay, and the resultant input and output blocks are periodic with respect to the sampled period (block size).
The Periodic Random signal is also an ergodic, stationary random signal consisting only of integer multiples of the FFT frequency increment. The frequency spectrum of this signal has random amplitude and random phase distribution. Following figure shows the spectrum characteristics.
Periodic Random Signal Spectrum
For each spectral average, input signal is generated with random amplitude and random phase. The system is excited with this input multiple blocks, until the transient response to the change in excitation signal decays. The input and response histories should then be periodic with respect the block-size and are saved as one spectra average in the total process. With each new average, a new block of signal, random with respect to previous input signals, is generated so that the resulting measurement will be completely randomized.
Also available as excitation waveforms are the sine type, i.e., Chirp, Burst Chirp.
Sometimes multiple excitation testing may be required, and this is referred as Multiple Input Multiple Output (MIMO) testing. It is more often used for large and supple structures’ modal testing. With use of multiple modal shakers, the excitation energy will be sufficiently distributed to excite the global modes of the structure under test. Simply increasing the driving force with single shaker arrangement, will overstress the driving point, and caused nonlinearity behavior of the structure.
Besides, for structures that have repeated modes, or highly coupled modes, MIMO modal testing will result in multiple columns of FRF. With this information, the repeated modes or highly coupled modes can be identified, using the corresponding FRF matrix. It needs to point out that the parameter identification method to handle the poly-reference FRF matrix will be the poly-reference type too. With identified mode participation factor, the repeated or highly coupled modes can be isolated.