Vibration analysis is a non-invasive condition monitoring technique used to detect mechanical problems in industrial machinery. It measures and analyzes the vibration signals produced by rotating or reciprocating equipment.
The technique helps maintenance teams identify, monitor, and prevent mechanical failures. As part of a predictive maintenance program, vibration analysis can detect developing problems before they cause equipment failure or unplanned downtime.
Vibration is the oscillatory movement of an object around a central equilibrium position. All rotating machines produce vibration during operation. Each machine also has its own characteristic vibration signature.
Changes in this vibration signature can indicate that a machine requires maintenance attention. Mechanical problems such as unbalance, misalignment, bearing defects, looseness, and gear wear can all change vibration patterns.
Vibration analysis measures these movements with specialized vibration sensors. Common sensors include accelerometers, velocity sensors, and displacement sensors.
Accelerometers measure changes in vibration acceleration and are widely used for machinery condition monitoring.
Velocity sensors measure vibration velocity. They are often used to assess the overall vibration and condition of rotating machinery.
Displacement sensors measure movement from a reference position. They are especially useful for low-frequency vibration and shaft measurements.
These sensors measure two important characteristics: vibration amplitude and frequency.
Vibration frequency is measured in Hertz (Hz) and describes the number of oscillations per second. Vibration amplitude measures the magnitude of the movement.
By examining these signals, analysts can identify patterns associated with specific machinery faults. These patterns appear in the frequency spectrum or time waveform.
Common vibration signatures can indicate:
- Unbalance: A dominant vibration around 1× shaft rotational speed.
- Misalignment: Vibration at 1× and 2× shaft speed, often with increased axial vibration.
- Bearing defects: High-frequency patterns associated with characteristic bearing defect frequencies.
- Mechanical looseness: Multiple harmonics and irregular vibration patterns.
- Gear faults: Sidebands around gear mesh frequencies.
- Resonance: Increased vibration near the natural frequency of a component.
- Rotor bar defects: Characteristic sidebands in induction motor vibration spectra.
Each fault produces a characteristic vibration pattern. These patterns help experienced analysts determine the underlying mechanical problem.
How does vibration analysis work?
Vibration analysis follows a systematic process. It transforms mechanical vibration into useful maintenance information.
The process typically includes five stages:
- Attach sensors to the equipment.
- Collect vibration data.
- Analyze frequency and amplitude.
- Identify irregularities.
- Generate a report and maintenance recommendations.
1. Attach sensors to the equipment
Correct sensor installation is essential for reliable vibration measurement.
Stud mounting provides a secure connection and a wide frequency response. Adhesive mounting offers a practical option for permanent installations. Magnetic mounting allows technicians to take measurements from suitable ferrous surfaces.
Sensors are usually positioned on bearing housings or structural locations close to rotating components.
Depending on the machine, measurements may be taken in radial and axial directions. The mounting surface should be clean and flat. Paint, corrosion, or other contamination can affect signal transmission.
2. Collect vibration data
Maintenance teams can collect vibration data in two main ways.
The first is route-based condition monitoring. Technicians use portable vibration analyzers to take periodic measurements from individual machines.
The second is continuous or online condition monitoring. Permanently installed sensors regularly collect data from the equipment.
Common measurements include:
- Overall vibration
- Frequency spectrum
- Time waveform
- Acceleration
- Velocity
- Displacement
Operating conditions are also important. Changes in machine speed or load can affect vibration. For this reason, measurements should be compared under similar operating conditions whenever possible.
3. Analyze frequency and amplitude
A Fast Fourier Transform (FFT) converts a time-domain vibration signal into a frequency spectrum.
The spectrum separates a complex vibration signal into individual frequencies. Analysts can then compare those frequencies with shaft speed, bearing frequencies, gear mesh frequencies, and other known machine characteristics.
Time waveform analysis provides another view of machine condition. Instead of looking at individual frequencies, it examines the original vibration signal over time.
This approach can reveal impacts, intermittent behavior, modulation, and other characteristics that may not be obvious from an overall vibration value.
Analysts can also use techniques such as envelope analysis or demodulation. These methods are particularly useful for detecting repetitive impacts associated with developing bearing faults.
Effective vibration analysis rarely depends on a single measurement. Analysts examine the relationship between frequency, amplitude, running speed, harmonics, sidebands, and waveform behavior.
4. Identify irregularities
The next step is to compare current vibration measurements with normal machine behavior.
Analysts may use:
- Baseline measurements
- Historical trends
- Alarm limits
- Previous measurements
- Known fault patterns
The objective is not simply to determine whether vibration is high or low. Analysts need to understand how the vibration has changed and what that change means for the machine.
For example, a developing bearing fault can produce characteristic vibration patterns before more obvious symptoms appear.
Early detection gives maintenance teams more time to investigate the problem and plan the appropriate maintenance action.
5. Generate reports
After analyzing the vibration data, the findings should be documented.
A vibration analysis report may include:
- Machine identification
- Measurement locations
- Operating conditions
- Vibration levels
- Historical trends
- Frequency spectra
- Time waveforms
- Identified faults
- Severity assessments
- Recommended maintenance actions
The purpose of vibration analysis is not simply to collect data. It is to turn condition data into actionable maintenance decisions.
When should you perform vibration analysis?
The correct monitoring frequency depends on several factors. These include asset criticality, machine condition, operating environment, failure behavior, and the P-F interval.
The P-F interval describes the period between a detectable potential failure and a functional failure.
Understanding this interval helps maintenance teams determine how often they should monitor an asset. Measurements must be frequent enough to detect deterioration while there is still time to respond.
Critical equipment usually requires closer monitoring. A failure on these machines may have a greater effect on production, safety, or maintenance costs.
Maintenance teams should also consider increasing the monitoring frequency when:
- Overall vibration begins to increase.
- New frequencies appear in the spectrum.
- An existing fault becomes more severe.
- Maintenance work could affect machine alignment or balance.
- Operating conditions change significantly.
Monitoring frequency should reflect how quickly a fault can develop rather than relying only on a fixed calendar schedule.
A quarterly measurement interval may not provide enough warning for machines with rapidly developing faults. Critical equipment may therefore require weekly measurements or continuous online condition monitoring.
Key parameters measured in vibration analysis
Three of the most important parameters in vibration analysis are acceleration, velocity, and displacement.
Each parameter provides a different view of machine vibration.
Acceleration
Acceleration measures how quickly vibration velocity changes over time.
Accelerometers offer a broad frequency response. They are particularly useful for detecting higher-frequency vibration.
This makes acceleration measurements useful for identifying conditions such as:
- Bearing faults
- Gear defects
- Mechanical impacts
- Pump cavitation
Velocity
Velocity measures the speed of vibrating movement.
Vibration velocity is widely used to assess the overall vibration severity of rotating machinery. It provides a useful representation of machine vibration across a broad frequency range.
For this reason, velocity is a common parameter in machinery condition monitoring.
Displacement
Displacement measures the distance a vibrating component moves from a reference position.
It is particularly useful for lower-frequency vibration and shaft measurements. Proximity probes are commonly used when monitoring shaft displacement on machinery with fluid-film bearings.
Vibration amplitude
Vibration amplitude can be expressed in several ways.
Peak-to-peak amplitude measures the total distance between the maximum positive and negative values. It is useful when mechanical clearance is important.
Peak amplitude measures the highest vibration level from the reference position.
Root Mean Square (RMS) represents the effective magnitude of a changing vibration signal. It is widely used when evaluating machinery vibration.
HBK provides further information about acceleration, velocity, displacement, frequency, and amplitude in its guide to measuring vibration.
Kurtosis and crest factor
Statistical parameters provide additional information about a vibration signal.
Kurtosis measures characteristics of the vibration amplitude distribution. Higher values can indicate impulsive behavior associated with impacts or mechanical faults.
Crest factor compares the peak amplitude with the RMS value. Changes in crest factor can help identify impulsive events associated with developing machinery faults.
What are the benefits of vibration analysis?
The main benefit of vibration analysis is early fault detection.
Traditional time-based maintenance relies on predetermined maintenance intervals. As a result, components may be replaced while they are still in acceptable condition. Faults can also develop between scheduled maintenance activities.
Condition-based maintenance and predictive maintenance use information about the actual condition of equipment to support maintenance decisions.
Reduce unplanned downtime
Unexpected machinery failure can interrupt production and cause secondary equipment damage.
Vibration analysis can identify conditions such as:
- Bearing deterioration
- Shaft misalignment
- Mechanical looseness
- Unbalance
- Gear defects
- Resonance
Detecting these problems earlier gives maintenance teams more time to plan repairs before the equipment reaches a critical condition.
Extend equipment service life
Excessive vibration increases mechanical stress.
Bearings, seals, couplings, gears, shafts, and other components can experience additional wear when machinery operates outside acceptable conditions.
Identifying and correcting the underlying problem can help machinery operate closer to its intended condition. This can support longer equipment service life.
Improve maintenance planning
Vibration analysis helps organizations move from reactive maintenance toward planned, condition-based maintenance.
Instead of treating every machine in the same way, maintenance teams can identify which assets require attention.
This supports better planning of:
- Maintenance resources
- Technician workloads
- Spare parts
- Repair schedules
- Production shutdowns
It also helps teams prioritize maintenance according to machine condition and criticality.
Improve safety
Mechanical failures can create safety risks, particularly on equipment operating at high speeds or under heavy loads.
Identifying developing problems before failure can reduce exposure to these risks.
Remote vibration monitoring can provide an additional benefit. Technicians can monitor certain assets without making frequent visits to hazardous or difficult-to-access locations.
Common applications of vibration analysis
Vibration analysis is widely used on rotating industrial equipment.
It is particularly valuable when machine reliability directly affects production, safety, maintenance costs, or product quality.
Pumps and motors
Pumps are used throughout manufacturing, water treatment, energy production, process industries, and building systems.
Vibration analysis can help identify:
- Cavitation
- Unbalance
- Misalignment
- Bearing deterioration
- Shaft problems
- Hydraulic disturbances
Electric motors can also develop mechanical problems that produce recognizable vibration patterns.
Monitoring these changes allows maintenance teams to investigate problems before they develop into serious failures.
Bearings and gears
Bearings and gearboxes can produce complex vibration signatures.
Rolling-element bearing faults can develop in the inner race, outer race, cage, or rolling elements. Different defects produce characteristic vibration frequencies.
Analysts often use envelope analysis to highlight repetitive impacts associated with bearing damage.
Gearbox vibration analysis can help identify:
- Gear tooth damage
- Gear wear
- Misalignment
- Bearing problems
- Abnormal gear mesh behavior
Turbines and compressors
Turbines and compressors often operate at high speeds. Small changes in vibration behavior can therefore provide important information about machine condition.
Potential vibration sources include:
- Unbalance
- Misalignment
- Resonance
- Critical-speed behavior
- Bearing problems
- Other dynamic machine conditions
These machines often benefit from continuous or frequent vibration monitoring because of their operational importance.
Fans and blowers
Industrial fans and blowers can experience vibration from several sources.
Common problems include:
- Impeller or wheel unbalance
- Material buildup
- Bearing deterioration
- Misalignment
- Bent shafts
- Blade-related vibration
- Structural resonance
Monitoring vibration trends allows maintenance teams to identify changes before they become more serious.
From vibration data to maintenance action
Vibration analysis provides the greatest value when condition data leads to maintenance action.
Collecting more data does not automatically improve reliability.
Maintenance teams need to understand what has changed, determine the likely cause, assess the severity, and decide what action is required.
This is where vibration analysis becomes more than a measurement technique.
When integrated into a broader condition monitoring and predictive maintenance strategy, it provides greater visibility into machinery health. It helps maintenance teams identify developing faults, prioritize maintenance, and make better-informed decisions.
The result is a shift from reacting to machine failures to identifying and addressing problems before they become critical.