【2026 Monitoring System】Bentley condition monitoring principle and calculation method for shaft vibration, installation clearance, and voltage
Thread Content
This post was last edited by The one on 2026-5-28 04:59. The Bently Nevada condition monitoring system is a mainstream solution for monitoring the vibration and shaft condition of industrial rotating machinery such as turbines, generators, compressors, pumps, etc. Its core principle relies on the real-time collection and analysis of key parameters such as mechanical vibration, shaft displacement, and rotational speed; by identifying characteristics of abnormal signals, it determines the health status of the equipment, enabling fault warning and diagnosis.
1. Core monitoring parameters and sensing principles: The system obtains physical quantities related to mechanical operation through various sensors, which are then converted into electrical signals for analysis. Vibration monitoring: Eddy current sensors (non-contact) or piezoelectric acceleration sensors (contact).
Eddy current sensor: Based on the principle of electromagnetic induction, the sensor probe creates an alternating electromagnetic field with the surface of the rotating shaft (metal). When the shaft vibrates, changes in the gap lead to variations in the intensity of the eddy current effect, which is converted into a voltage signal proportional to the vibration displacement (it is typically used to measure relative shaft vibration, with accuracy at the micron level).
Its \"force-to-electricity conversion\" property transforms the acceleration of mechanical vibrations into a charge signal, which, after amplification, produces a voltage signal related to the intensity of the vibration (usually measuring the absolute vibration of the housing).
Shaft displacement/thickness variation monitoring: This relies primarily on eddy current sensors, which measure the axial position changes of the rotating shaft (such as shaft play) or the relative expansion between the rotor and stator (thickness variation), thereby generating a linear voltage signal that indicates the axial stability of the shaft system (to prevent friction between moving and stationary components).
Speed and phase monitoring: Utilizes magnetoelectric sensors or photoelectric sensors. Magnetoelectric sensor: It calculates the rotational speed by detecting the pulse signals generated when the gears/keys on the rotating shaft cut through the magnetic field (the pulse frequency is proportional to the rotational speed).
Phase sensor (synchronization signal): Collected in synchronization with the vibration signal, it is used to analyze the vibration phase and identify the location of faults (such as the phase characteristics corresponding to imbalance or misalignment).
2. Signal Processing and Feature Extraction: The raw signals collected by sensors (such as vibration and displacement) are amplified and filtered using pre-amplifiers before being transmitted to the monitoring host (such as 3500, 1770 series frameworks). Fault features are extracted through the following processes: Time-domain analysis: The peak value, root mean square (RMS) value, and peak-to-peak value of vibrations are calculated to determine whether the vibration intensity exceeds the standard thresholds (such as ISO 10816). Frequency domain analysis: Converts a time-domain signal into a spectrum using the Fourier Transform (FFT), in order to identify characteristic frequencies such as the rotation frequency f, 2f, harmonics, etc.
For example: rotor imbalance corresponds to vibration peaks at the f frequency； Misalignment corresponds to the 2f frequency peak ； Bearing failures correspond to specific frequencies (for example, the frequency of inner-ring failure = 0.6f × number of bearing balls).
Trend analysis: Long-term recording of parameter variation curves (such as the trend of vibration levels over time) allows the rate of equipment degradation to be determined by changes in slope (for example, a sudden increase in vibration levels may indicate increased bearing wear).
3. Fault diagnosis and early warning logic: The system enables hierarchical early warning and diagnosis based on preset thresholds (alarm values, critical values) and a fault characteristic database. Threshold alarm: When the vibration value or displacement value exceeds the set threshold (e.g., a value corresponding to an ‘attention’ level triggers an alarm, while a value indicating a dangerous level leads to shutdown), the system activates audible and visual alarms and records the time at which this occurs. Feature matching: The real-time spectral features are compared with a database of typical fault characteristics (such as imbalance, misalignment, shaft bending, oil film vibration, etc.), to automatically or assistively determine the type of fault (for example, if a first harmonic appears in the spectrum with a stable phase, it is likely due to imbalance). Interlocked protection: For critical equipment such as turbines, when parameters reach dangerous levels, the system can generate interlock signals to shut down the equipment automatically, thereby preventing catastrophic accidents such as shaft breakage or fires caused by static and dynamic friction.