Introduction
Condition monitoring is the practice of monitoring parameters that determine the operating condition of machinery. This process predicts impending equipment failure by noting significant changes in machinery behavior. Condition monitoring is a significant component of predictive maintenance, where scheduled maintenance can be used to prevent equipment failure, before it occurs. The prediction of the time of failure is a cost-effective means of maintaining rotating machinery and stationary machinery, such as boilers and heat exchangers.
Vibration Monitoring in Rotating Machinery
Monitoring vibration in rotating machinery is an important aspect of condition monitoring. The vibration signatures in rotating machinery can be complex, so specialized training and experience is necessary to interpret the data. Today’s technology, however, automatically provides the analyst with the necessary parameters to allow assessment of equipment vibratory behavior. Analyzing vibration data is typically a matter of interpreting the frequency content and amplitude of the vibration signal. The vibratory frequencies typically can be interpreted in lieu of mechanical component performance. These components include rolling element bearings and out-of-balance or misaligned shafts.
Analyzing the Vibration Frequency Content
Vibration frequencies and amplitudes usually tell a story. For example, frequencies that correspond to the equipment’s rotational speed usually indicate an imbalance that can be corrected by rebalancing the machine. Rolling element bearings usually exhibit increasing vibratory amplitudes at specific frequencies. These frequencies usually indicate a bearing that is wearing out. When this occurs, the bearing can be replaced, before total equipment failure occurs. This type of maintenance activity is very cost-effective, as the cost of machinery down time can be very large.
Analyzing the Vibration Amplitude
Today’s vibration data collectors utilize a Fast Fourier Transform (FFT) to convert the time domain vibration signal into a frequency domain signal. Utilizing the frequency content (spectral analysis) is only one method of interpreting machinery vibration data. Depending on the equipment type, a maintenance analyst may also use the time domain data, historical trends, the shape of the vibration signal, and the phase relationships between various amplitudes.
Measuring Vibration with Handheld Data Collectors
Advances in technology and computer software now allow the technician to take measurements using handheld data collectors. This greatly simplifies the data collection process, and can allow for data analysis on site. Larger equipment may still require a dedicated vibration monitoring system, but many equipment types can be analyzed using handheld data collectors. Using either method of data collection, the diagnostic tools are generally the same.

Introduction
Vibration measurements have been used to reliably diagnose performance problems in machinery and related mechanical products. A vibration data collector can be used effectively to measure and analyze the machinery vibration content in gearboxes, engines, turbines, fans, compressors, pumps and bearings. Ideally, a machine will have little or no vibration, indicating that the rotating components are appropriately balanced, aligned, and well maintained. Quick analysis and assessment of the vibration content can lead to fault diagnosis and prognosis of a machine’s ability to continue running. When compared to historical vibration data, vibration measurements can be used to pinpoint mechanical defects such as unbalance, misalignment, resonance, and part loosening.

Collecting Vibration Data
During vibration condition monitoring, the data from the rotating machinery is collected, stored, interpreted, and then compared with historical vibration measurements. Depending on the machine, the vibration data may be gathered using two different techniques. Vibration measurements on larger machines are taken using permanently attached transducers (accelerometers). For smaller machines (with hard-to-reach locations), vibration data is taken using handheld data collectors with portable transducers. The data is then stored and analyzed to determine the machine’s mechanical condition.

Analyzing Vibration Data
Vibration measurements can be expressed in terms of displacement, velocity, acceleration, and high frequency content (for bearing condition detection). Most data collectors use FFT (Fast Fourier Transform) to convert the data from the time domain to the frequency domain. Vibration data collectors have a built-in PC interface that allows transfer of the measurements to a PC for data management. Analysis software displays spectrum, trend, waterfall plots, and waveform for advanced analysis. To assess a machine, the vibration data is compared with historical profiles from the same machine.

Vibration Data Collector Features
Quick, accurate diagnosis of machine vibration conditions, including bearing vibration
Simplifies the task and increases the speed of collecting vibration monitoring data
Incorporates the latest software for vibration data analysis and performance prediction
Combines basic analysis functions and rotating machinery balancing capabilities
Lightweight, compact portable vibration data collectors are easy to use
Used with a portable computer, such as a laptop or notebook
Multiple channel models are available
Usually do not require any special training