Random vibration experiments typically involve a reference curve, known as the Power Spectral Density (PSD) pattern, which defines the distribution of vibration energy across different frequencies. The solid line in the figure represents this reference PSD, which is specified to be within the frequency range of 50–1000 Hz. The power spectrum of the signal applied to the vibration table is designed to be flat, indicating uniform energy distribution. The solid line also shows the measured power spectrum from the experiment, referred to as the control spectrum. The upper and lower dotted lines represent the tolerance limits, set at +3 dB and -3 dB, respectively, to ensure that the actual vibration remains within acceptable bounds. Each test session is structured with specific durations: 20 seconds at -6 dB and -3 dB levels before reaching the main test condition at 0 dB, which lasts for 60 seconds. Since the input energy to the vibration table during broadband random testing is very low, the power spectral density is only 0.0025 g²/Hz between 50 and 1000 Hz. Therefore, the structural response is considered a stationary random vibration. This type of vibration can be modeled using linear systems, and the dynamic behavior of the support before and after bolt loosening can be analyzed through power spectrum techniques.
A spectrally reproducible wideband digital random vibration control system generally consists of four main components: A/D and D/A conversion, time-frequency domain transformation, spectrum equalization, and excitation signal randomization. One key challenge in random vibration testing is accurately calculating the Power Spectral Density (PSD). In theory, an infinite-length time record would provide the most accurate representation of a random process. However, in practice, only a finite-length data segment, or "sample," is available. This sample contains all the statistical information of the entire process. To compute the PSD, the experiment employs the Welch method, which involves windowed segmentation and averaging of periodograms to improve accuracy and reduce variance.
Another critical aspect is achieving spectrum equilibrium during the test. While a random signal source can be used to drive the vibration table, the actual response depends on the dynamic characteristics of the system, including the test specimen and the table itself. As a result, it can be challenging to match the required reference spectrum, Sr(ω), specified in the experimental protocol.
The experimental setup includes measuring the acceleration responses at multiple points (points 2, 3, and 4) under normal bolt tightening conditions. Point 1 is used to control the vibration table. After loosening the bolts in the reverse direction, the clamping force is reduced, increasing the gap between the straps to simulate varying degrees of bolt looseness. The same testing procedure is repeated, and the resulting data is recorded for analysis.
Economic American Type Scaffolding Casters
Economic American Type Scaffolding Casters
Economic American Type Scaffolding Casters
Ningbo Mywin Caster Co., Ltd. , https://www.mywin-caster.com