Vibration analysis is a condition monitoring technique that measures, records, and analyzes the vibration signatures of rotating and reciprocating equipment to detect developing mechanical faults before they produce functional failure. Every rotating machine produces a characteristic vibration signature — a combination of frequencies, amplitudes, and waveform patterns determined by its rotational speed, component geometry, and mechanical condition. When a fault develops — a bearing defect initiates, a shaft becomes misaligned, a gear tooth cracks — the vibration signature changes in specific, identifiable ways. Vibration analysis detects and interprets those changes.
Vibration analysis is a form of non-destructive testing (NDT) — it monitors equipment condition without stopping production or disassembling the asset. Measurements are taken with the equipment running under normal operating conditions, producing condition data that reflects the actual in-service state of the machine rather than a snapshot taken during a shutdown inspection. This makes vibration analysis one of the most operationally non-intrusive condition monitoring techniques available.
As a condition monitoring technique, vibration analysis feeds Condition-Based Maintenance (CBM) and Predictive Maintenance (PdM) programs by providing the earliest available warning of developing faults on rotating equipment. For bearing failures specifically — one of the most common rotating equipment failure modes — vibration analysis can detect defects weeks or months before they progress to functional failure, providing ample time for planned corrective maintenance.
Why Vibration Analysis Matters
Bearing failures, misalignment, imbalance, looseness, and gear faults collectively account for the majority of rotating equipment failures in industrial operations. These failure modes share a critical characteristic: they develop gradually, producing progressively worsening vibration signatures as they advance. The interval between detectable onset and functional failure — sometimes weeks, sometimes months — is the window that vibration analysis exploits.
Without vibration monitoring, these failures are typically detected when they have progressed to audible noise, elevated temperature, or visible damage — late in the failure progression, with little remaining time for planned response. With vibration monitoring, the same failure is detected at its earliest measurable stage, when planned replacement can be scheduled at a convenient maintenance window with parts staged and the right technicians assigned.
The financial case is direct. A planned bearing replacement on a critical pump — scheduled based on a vibration analysis finding — costs a fraction of the same repair performed as an emergency response to an unexpected failure. Multiply that differential across the rotating equipment population of an industrial facility and vibration monitoring programs consistently deliver positive returns through reduced emergency repair costs, eliminated production losses, and extended component life from optimized replacement timing.
How Vibration Analysis Works
Measurement
Vibration measurements are taken using sensors placed at defined measurement points on the equipment — typically on bearing housings, gearbox casings, and motor frames where vibration transmission from internal components is highest. Three sensor types are used depending on application requirements:
- Accelerometers: The most common vibration sensor type. Measure acceleration in units of g (gravitational acceleration) or m/s². Best for detecting high-frequency bearing defect frequencies and structural resonance. Available in handheld route-based configurations and permanently mounted online monitoring installations.
- Velocity sensors: Measure vibration velocity in mm/s or in/s. Velocity is the standard measurement parameter for overall vibration severity assessment per ISO 10816 and related standards. Most sensitive in the mid-frequency range relevant to imbalance, misalignment, and looseness.
- Displacement sensors (proximity probes): Measure shaft displacement directly, typically installed in journal bearing systems on large turbomachinery. Provide shaft orbit and centerline position data that accelerometers and velocity sensors cannot capture.
Analysis Methods
Raw vibration signals are processed using several analytical techniques to extract fault information:
Fast Fourier Transform (FFT) spectrum analysis converts the time-domain vibration signal into a frequency-domain spectrum showing the amplitude of vibration at each frequency component. Fault frequencies — bearing defect frequencies, gear mesh frequencies, imbalance at running speed, misalignment at harmonics of running speed — appear as peaks at specific, calculable frequencies in the spectrum. FFT analysis is the foundation of most vibration fault diagnosis.
Waveform analysis examines the time-domain signal directly, revealing impact patterns, modulation characteristics, and waveform shape that indicate specific fault types. Bearing defects produce characteristic impact patterns in the waveform that may be present before they are clearly visible in the FFT spectrum.
Envelope analysis (demodulation) is specifically effective for early-stage bearing defect detection. It extracts the low-frequency impact repetition rate from high-frequency vibration signals, enabling detection of bearing defects at stages too early to appear clearly in standard FFT analysis.
Overall vibration level trending tracks the total vibration amplitude over time, providing a simple indicator of overall machine condition change. ISO 10816 defines severity zones for overall vibration levels by machine class — alert and danger thresholds can be set based on these standards and asset-specific baselines.
What Vibration Analysis Detects
Vibration analysis detects the following fault categories on rotating equipment:
- Bearing defects: Defects on inner race, outer race, rolling elements, and cage produce characteristic defect frequencies calculable from bearing geometry and rotational speed. Early-stage bearing defects are detectable through envelope analysis before they progress to advanced stages visible in standard FFT spectra. See: Spalling.
- Misalignment: Angular and parallel misalignment between coupled shafts produces characteristic vibration at harmonics of running speed, typically with elevated axial vibration relative to radial vibration.
- Imbalance: Mass imbalance in rotating components produces vibration at running speed (1x), with amplitude proportional to the degree of imbalance and the square of rotational speed.
- Looseness: Mechanical looseness — loose bearing fits, loose foundation bolts, loose rotor components — produces vibration at sub-harmonics and harmonics of running speed with characteristic truncated waveform patterns.
- Gear faults: Tooth wear, cracking, and spacing errors produce vibration at gear mesh frequency (number of teeth times rotational speed) and its harmonics, with sidebands indicating the severity and distribution of tooth damage.
- Resonance: Structural or component resonance occurs when a forcing frequency coincides with a natural frequency, producing amplified vibration that can cause accelerated fatigue. Vibration analysis identifies resonance conditions and supports operational or design changes to detune them.
Vibration Analysis by Industry
Manufacturing: Vibration monitoring programs in manufacturing focus on production-critical rotating assets — motors, gearboxes, pumps, fans, and conveyor drives — where unplanned bearing or mechanical failures stop production. Route-based vibration measurement programs collect data on critical assets at defined intervals, with findings generating corrective work orders for planned bearing replacements and alignment corrections. In high-speed precision machining, vibration analysis also monitors spindle bearing condition, where early-stage defects affect part quality before they produce audible noise or machine shutdown.
Mining: Vibration analysis in mining addresses high-value, high-consequence rotating equipment — crusher bearings, mill drive gearboxes, conveyor drive systems, and large pump installations — where failures shut down processing circuits and produce significant production losses. Online vibration monitoring systems on critical fixed plant equipment provide continuous surveillance and automatic alerts, reducing reliance on periodic route-based measurements for the most critical assets. For mobile mining equipment, vibration analysis is applied to drivetrain components during scheduled maintenance inspections.
Oil and Gas: Online vibration monitoring is standard on critical rotating equipment in oil and gas — compressors, pumps, and turbines — providing both condition monitoring and protective shutdown capability. API 670 defines the standard for machinery protection systems using vibration and position monitoring on critical turbomachinery. Vibration trending on compressor and pump trains detects developing bearing, seal, and rotor faults before they progress to failures that could trigger process upsets or safety incidents.
Crane and Rigging: Vibration analysis on crane hoist gearboxes, slewing ring drives, and travel wheel assemblies detects bearing and gear faults before they produce failures under load. For large cranes operating in environments where access for physical inspection is limited — offshore cranes, tower cranes, large overhead cranes in active facilities — vibration monitoring reduces the frequency of required intrusive inspections while maintaining detection coverage for safety-critical mechanical components.
Common Vibration Analysis Program Failures
Inconsistent measurement points and technique: Vibration data is only trended meaningfully when collected at the same measurement point, in the same direction, with the same sensor type and mounting method, at consistent intervals. Variation in any of these factors produces data scatter that masks real condition changes. Marked measurement points, documented routes, and consistent measurement technique are prerequisites for reliable trending.
No baseline established before trending begins: Alert thresholds and fault diagnosis both require a known baseline — the normal vibration signature of the equipment in good condition at normal operating speed and load. Programs that begin monitoring without establishing baselines have no reference against which to evaluate subsequent measurements and cannot reliably distinguish normal variation from developing faults.
Overall level monitoring without spectrum analysis: Overall vibration level trending detects that something has changed but cannot identify what has changed or why. Spectrum analysis is required to diagnose fault type and severity. Programs that rely exclusively on overall level thresholds miss early-stage faults that change the frequency signature before they raise overall amplitude, and cannot provide the fault-specific diagnosis needed to plan appropriate corrective action.
Findings not converted to work orders: A vibration analysis finding that does not generate a corrective work order within the available P-F interval has not prevented a failure — it has documented one in advance. The process for converting analyst findings to work orders must be fast, reliable, and tracked. Findings that sit in analysis reports without triggering maintenance action provide no operational benefit.
Monitoring interval longer than P-F interval: If the monitoring interval exceeds the P-F interval for the failure mode being monitored, failures will progress to functional failure between measurements. Bearing P-F intervals on high-speed equipment can be as short as two to four weeks. Monthly measurement intervals on such equipment may miss failures entirely. Monitoring frequency should be matched to the failure mode’s P-F interval, with higher-criticality assets measured more frequently.
Vibration Analysis vs. Related CM Techniques
- Vibration analysis: Detects mechanical faults in rotating and reciprocating equipment through vibration signature measurement and analysis. Most effective for bearing defects, misalignment, imbalance, looseness, and gear faults. Requires equipment to be running during measurement.
- Oil analysis: Detects wear, contamination, and lubricant degradation through lubricant sample analysis. Complements vibration analysis — oil analysis detects wear metal accumulation from bearing and gear faults, providing confirmation and additional failure mode coverage. See: Oil Analysis / Lubricant Analysis.
- Thermography: Detects thermal anomalies from electrical faults, mechanical friction, and insulation breakdown. Covers failure modes not detectable by vibration analysis, particularly electrical faults in motors and switchgear.
- Ultrasonic testing: Detects high-frequency emissions from early-stage bearing defects, leaks, and electrical discharge. Can detect bearing defects earlier than standard vibration analysis in some applications, making it a useful complement for critical bearing monitoring programs.
- Motor circuit analysis: Detects electrical faults in motor windings and rotor bars. Covers motor failure modes that vibration analysis detects only at advanced stages.
Frequently Asked Questions
What faults does vibration analysis detect?
Vibration analysis detects bearing defects (inner race, outer race, rolling element, and cage faults), misalignment between coupled shafts, mass imbalance in rotating components, mechanical looseness in bearing fits and structural connections, gear tooth faults (wear, cracking, spacing errors), and structural resonance conditions. It is most effective for rotating and reciprocating machinery — motors, pumps, fans, gearboxes, compressors, and conveyor drives. It does not directly detect electrical faults, lubricant condition, or corrosion, which require complementary techniques.
How often should vibration measurements be taken?
Measurement frequency should be matched to the P-F interval of the failure modes being monitored and the criticality of the asset. For most industrial rotating equipment, monthly route-based measurements provide adequate coverage for bearing and mechanical fault detection. For high-speed equipment or assets where bearing P-F intervals are short, bi-weekly or weekly measurements may be required. Critical assets in oil and gas, power generation, and continuous process industries typically use online continuous monitoring rather than periodic route measurements, providing real-time detection without interval limitations.
What is the difference between online and route-based vibration monitoring?
Route-based monitoring uses portable instruments carried by a technician who visits each measurement point at defined intervals — typically monthly or quarterly. It provides periodic snapshots of equipment condition and is cost-effective for monitoring large asset populations. Online monitoring uses permanently installed sensors connected to a continuous data acquisition system, providing real-time vibration data and automatic alert generation. Online monitoring eliminates the interval limitation of route-based programs and is appropriate for critical assets where continuous surveillance is required. Many programs combine both approaches — online monitoring on the most critical assets and route-based monitoring on the broader equipment population.
How does a CMMS support vibration analysis programs?
A CMMS supports vibration analysis by scheduling measurement routes, tracking route completion against defined intervals, storing measurement data against asset records for trending, and generating corrective work orders when findings exceed alert thresholds or indicate developing faults. When vibration findings and corrective work orders are stored in the same asset record, the CMMS builds the failure history that supports root cause analysis, PM interval validation, and reliability improvement tracking over time.
Related Terms
- Condition Monitoring (CM)
- Condition-Based Maintenance (CBM)
- Predictive Maintenance (PdM)
- Oil Analysis / Lubricant Analysis
- Spalling
- Mean Time Between Failures (MTBF)
- Non-Destructive Testing (NDT)
Build a Vibration Monitoring Program With Redlist
Redlist connects vibration measurement routes, findings, and corrective work orders in one platform — giving reliability teams the workflow to detect developing faults and act before failures occur.
