Vibration Analysis
How does FFT distinguish between imbalance and misalignment?
Imbalance generates a dominant 1× RPM (running speed) peak in the frequency spectrum; misalignment typically shows high 1× and 2× RPM peaks, often with axial vibration dominance.
What is the difference between time-domain and frequency-domain analysis?
Time-domain (waveform) shows overall amplitude and impact severity (e.g., bearing defects); frequency-domain (spectrum) identifies the specific source of vibration by isolating frequencies.
Why is VA more sensitive than temperature monitoring?
Vibration changes occur weeks or months before temperature rises, allowing detection of early-stage degradation that thermal sensors miss.
Vibration Analysis (VA) is a dominant condition monitoring technique in predictive maintenance that measures and analyzes oscillatory motions (frequency and amplitude) of rotating machinery to detect early mechanical faults like imbalance, misalignment, and bearing wear before catastrophic failure occurs. It uses sensors like accelerometers and signal processing such as Fast Fourier Transform (FFT) to convert raw data into frequency spectra for fault identification.
In shop floor maintenance, Vibration Analysis is applied through data acquisition using accelerometers on assets like pumps, motors, and fans, either via route-based collectors or wireless real-time monitoring. Signal processing via Fast Fourier Transform (FFT) converts raw time-domain waveforms into frequency spectra to isolate specific fault signatures. New readings are compared against established 'healthy' baselines to detect minor deviations, often weeks or months before visible symptoms appear. In CMMS/Asset Reliability platforms like ServiceGrid, VA data triggers automated work orders, tracks asset health trends, and correlates findings with thermography or oil analysis for root cause validation.
Ignoring Baselines: failing to record initial 'healthy' spectra leads to inability to distinguish normal operational variance from developing faults.
Delayed Intervention: Continuing to run equipment after detecting abnormal amplitude increases results in bearing spalling, shaft cracks, or catastrophic seizure.
Inadequate Sensor Placement: Mounting sensors on non-structural locations or loose surfaces distorts data, masking critical faults like misalignment or resonance.