What Vibration Analysis Is
Every rotating machine produces mechanical vibration when it operates.
This is not a sign of malfunction.
It is a physical consequence of rotating mass, bearing contact forces, gear mesh interactions, and structural resonances that are present in every motor, pump, fan, and gearbox regardless of their condition.
The vibration a machine produces in healthy condition has a characteristic signature — a specific amplitude and frequency distribution that reflects the machine's geometry, speed, and load.
When a fault begins to develop — a bearing defect, a gear tooth problem, a shaft imbalance, a misalignment — the fault produces a change in the vibration signature.
A bearing with a developing outer race defect produces impacts at a specific frequency determined by the bearing geometry and shaft speed — the ball pass frequency outer race (BPFO) — that is not present in the healthy bearing's vibration signature.
A gear with a developing tooth defect produces vibration at the gear mesh frequency and its harmonics with a characteristic amplitude modulation pattern that differs from a healthy gear pair.
A shaft that has developed imbalance produces increased vibration amplitude at the shaft running speed — the fundamental frequency — that is not present when the shaft is balanced within tolerance.
Vibration analysis is the practice of measuring these vibration signatures, comparing them against healthy baselines, and identifying the specific fault frequencies that indicate developing failure modes — before those failure modes progress to functional failure that stops production and requires emergency repair.
Why Vibration Analysis Matters for Manufacturing
The financial case for vibration analysis in manufacturing maintenance is built on a single consistent observation.
Bearing failures — the most common rotating equipment failure mode in manufacturing — develop over weeks to months in a way that produces clear vibration signature changes before functional failure occurs.
The P-F interval for a typical rolling element bearing failure — the time between the first detectable vibration anomaly and the functional failure that stops the machine — is four to eight weeks in most industrial applications.
A manufacturing operation without vibration monitoring on its critical rotating equipment discovers bearing failures when the bearing produces the sudden, loud, and production-stopping failure that marks the end of that four-to-eight-week window.
A manufacturing operation with vibration monitoring discovers the developing bearing fault at the beginning of that window — and replaces the bearing during a planned maintenance interval within the next two to three weeks, before functional failure occurs.
The difference in outcome between these two scenarios is the entirety of the case for vibration analysis investment.
In the first scenario: an unplanned production stop of 90 to 240 minutes, emergency parts procurement, overtime labor, potential secondary damage to the shaft and housing, and a reactive repair that costs three to four times more than a planned replacement.
In the second scenario: a planned 45-minute bearing replacement during a production window that was accommodated in the schedule, standard parts from confirmed storeroom stock, regular labor, and no secondary damage.
Multiplied across a manufacturing facility's critical rotating equipment fleet — typically 50 to 200 motors, pumps, and gearboxes on a mid-sized production site — vibration analysis prevents the bearing failures that collectively account for a significant proportion of unplanned downtime and maintenance cost.
The Physics of Vibration Measurement
Understanding what vibration analysis measures requires a basic understanding of the physical quantities involved — not at an engineering depth, but at the level that allows maintenance and operations professionals to evaluate implementation options and interpret summary results.
What is measured
Vibration is measured as the oscillation of a physical structure around its rest position.
Three physical quantities describe this oscillation.
Displacement describes how far the structure moves from its rest position — measured in micrometers or thousandths of an inch. Displacement measurement is most relevant for low-frequency vibrations — shaft orbit analysis in journal bearings and machinery alignment assessment.
Velocity describes how fast the structure is moving — measured in millimeters per second or inches per second. Velocity measurement is most commonly used for general machine condition assessment because it provides a consistent measure across a wide frequency range.
Acceleration describes how rapidly the velocity is changing — measured in meters per second squared or in g (multiples of gravitational acceleration). Acceleration measurement is most sensitive to high-frequency vibrations — the impact events associated with bearing defects and gear tooth faults.
How it is measured
Vibration is measured using accelerometers — electronic sensors that convert mechanical vibration into an electrical signal proportional to the acceleration of the surface they are attached to.
Piezoelectric accelerometers are the most widely used type in industrial condition monitoring — they are robust, accurate across a wide frequency range, and available in a variety of form factors for different installation requirements.
The electrical signal from the accelerometer is processed by a data collector or monitoring system — converting the raw time-domain signal into frequency-domain data through a mathematical transformation called the Fast Fourier Transform (FFT) that reveals the specific frequencies present in the vibration signature.
How faults are detected
In a healthy machine, the vibration frequency spectrum contains peaks at the fundamental operating frequencies — shaft running speed and its harmonics, gear mesh frequencies, bearing pass frequencies — with amplitudes characteristic of healthy operation.
When a fault develops, new frequency components appear in the spectrum — at the frequencies specific to the fault type — or existing frequency components increase in amplitude beyond their healthy baseline levels.
The vibration analyst identifies these anomalous frequency components, compares them against the known fault frequency signatures for the specific machine's geometry, and diagnoses the developing fault type from the pattern of anomalous frequencies.
Vibration Fault Signatures: What Different Faults Look Like
Each rotating equipment fault type produces a characteristic vibration signature — a pattern of frequency components and amplitude behaviors that distinguishes it from other fault types and from healthy machine vibration.
Understanding these signatures at a basic level is useful for maintenance and operations professionals evaluating vibration analysis results and communicating with specialists.
Bearing faults
Rolling element bearing faults produce vibration at specific frequencies calculated from the bearing geometry — the bearing dimensions and the number of rolling elements — multiplied by the shaft speed.
Four bearing fault frequencies are calculated for every bearing.
The Ball Pass Frequency Outer Race (BPFO) is excited when a rolling element contacts a defect on the outer bearing race.
The Ball Pass Frequency Inner Race (BPFI) is excited when a rolling element contacts a defect on the inner bearing race.
The Ball Spin Frequency (BSF) is excited when a defect is present on the rolling element surface itself.
The Fundamental Train Frequency (FTF) is excited by cage defects.
Early-stage bearing faults produce impulses at these frequencies with amplitudes that are small relative to the overall vibration level — requiring acceleration measurements and signal processing techniques that enhance the impulsive components relative to the background vibration.
As the fault progresses, the amplitude of the fault frequency components increases, and the overall vibration level rises as the bearing condition deteriorates.
Imbalance
Rotating imbalance produces vibration primarily at the shaft running speed — the fundamental frequency — in the radial direction.
The amplitude is proportional to the amount of imbalance and increases with the square of the rotational speed.
Imbalance is one of the most common sources of elevated vibration in rotating machinery and one of the easiest to diagnose from the vibration spectrum — a dominant peak at the fundamental frequency with no corresponding peaks at harmonic frequencies is the classic imbalance signature.
Misalignment
Shaft misalignment between a motor and pump, for example, produces vibration predominantly at the shaft running speed and its second harmonic — with the relative amplitude of the fundamental and second harmonic depending on whether the misalignment is primarily angular, parallel, or a combination.
Misalignment also produces elevated axial vibration — vibration in the direction of the shaft axis — that helps distinguish it from imbalance, which is primarily radial.
Gear faults
Gear faults produce vibration at the gear mesh frequency — the shaft speed multiplied by the number of gear teeth — and its harmonics.
A gear tooth defect produces sidebands around the gear mesh frequency — additional frequency components spaced at shaft running speed intervals above and below the mesh frequency — with amplitudes that increase as the tooth defect progresses.
Gear fault diagnosis from vibration spectra requires knowledge of the gear geometry — specifically the number of teeth on each gear — to calculate the expected mesh frequencies and identify anomalous sideband patterns.
Two Implementation Approaches
Vibration analysis is implemented through two fundamentally different approaches — each with different cost profiles, different detection capabilities, and different appropriate applications.
Periodic route-based measurement
In route-based measurement, a trained analyst or technician carries a portable vibration data collector to each measurement point on a defined route — collecting vibration measurements from each point at each visit and comparing the measurements against historical baselines and alarm thresholds.
Route-based measurement is the most widely deployed approach for the general rotating equipment fleet — it is cost-effective because one portable instrument covers dozens or hundreds of measurement points rather than requiring a permanently installed sensor at each point.
The measurement interval — weekly, monthly, or quarterly — determines the minimum detectable P-F interval. A monthly measurement interval can detect faults with P-F intervals of four weeks or longer. It cannot detect faults that develop and progress to functional failure between measurement visits.
Route-based measurement requires trained personnel to collect data reliably — consistent measurement technique, correct sensor placement, and correct instrument settings — and either trained analysts or automated diagnostic software to interpret the results.
Online continuous monitoring
In online monitoring, permanently installed sensors at each measurement point transmit vibration data continuously to a central monitoring system — detecting threshold crossings and trend changes in real time rather than between periodic measurement visits.
Online monitoring is most appropriate for the highest-criticality assets where continuous monitoring is justified by the failure consequence — Tier 1 assets whose failure stops production entirely and whose P-F interval is short enough that monthly route-based measurement might miss developing faults.
The cost of online monitoring — permanently installed sensors, cabling or wireless transmission infrastructure, and monitoring system licensing — is significantly higher than route-based measurement per monitoring point.
But for the ten to twenty most critical rotating assets in a manufacturing facility, the production loss prevention value of continuous monitoring typically justifies the investment — because a single prevented failure on a critical production asset recovers the sensor infrastructure cost many times over.
The practical implementation for most manufacturing facilities combines both approaches — online monitoring for the five to ten highest-criticality rotating assets with short P-F intervals, and route-based monitoring for the broader fleet.
Vibration Analysis and the Maintenance Response
Vibration analysis only delivers its full financial value when the faults it detects are addressed through planned maintenance interventions within the remaining P-F interval.
This is the action gap problem — the time between vibration analysis detection and a planned maintenance intervention determines whether the detection prevents the failure or simply provides earlier warning of a failure that happens anyway.
A vibration analysis program that generates fault detection reports reviewed in a monthly maintenance meeting — where a bearing fault detected three weeks ago is first discussed and a work order created — has consumed most of the P-F interval in administrative delay before any maintenance action begins.
A vibration analysis program that automatically generates a maintenance work order when a configured fault frequency amplitude threshold is crossed — delivering the work order to the responsible technician's mobile device with the bearing identification, the recommended replacement procedure, and the parts availability confirmation — acts within hours of detection rather than weeks.
The difference between these two scenarios is the connection between the condition monitoring system and the maintenance execution system.
When they are separate — the vibration monitoring platform in one system, the CMMS in another — the connection depends on human intermediation that introduces latency and inconsistency.
When they are integrated — the vibration detection triggering automatic work order generation in the same platform — the action gap is eliminated and the full P-F interval is available for planning and execution rather than being consumed by administrative coordination.
This integration — between condition monitoring detection and maintenance execution response — is what converts vibration analysis from an expensive awareness tool into a genuine failure prevention program.
Frequently Asked Questions
What equipment should be monitored with vibration analysis?
The priority candidates for vibration analysis are rotating equipment assets that meet three criteria.
The asset is Tier 1 or Tier 2 in the criticality classification — its failure produces significant production, safety, or quality consequences.
The dominant failure mode produces vibration-detectable precursor signals — bearing wear, imbalance, misalignment, and gear faults all meet this criterion.
The P-F interval for the dominant failure mode is long enough for a planned maintenance response — typically two weeks or longer for most industrial bearing applications.
In a typical manufacturing facility, this includes critical pump motors, primary drive motors on production lines, gearboxes on high-speed production equipment, fans and compressors in production-critical HVAC and process systems, and centrifugal separators and mixers in process manufacturing.
How often should route-based vibration measurements be taken?
Measurement frequency should be matched to the P-F interval of the dominant failure modes for each asset.
A bearing failure mode with a P-F interval of four to six weeks warrants monthly measurement — providing at least two measurement opportunities within the P-F interval for detecting the fault before functional failure.
A bearing failure mode with a P-F interval of eight to twelve weeks can be adequately monitored with quarterly measurement.
Assets with known short P-F intervals — where historical failures have progressed rapidly — warrant more frequent measurement or online continuous monitoring.
Does vibration analysis require specialist expertise?
Route-based data collection can be performed by maintenance technicians trained in consistent measurement technique — correct sensor placement, correct instrument settings, and consistent measurement conditions.
Vibration spectrum analysis and fault diagnosis require either a trained vibration analyst with certification from a recognized body such as the Vibration Institute or the ISO 18436-2 standard, or an automated diagnostic system that applies fault frequency analysis algorithms without requiring manual spectrum interpretation.
Most manufacturing operations start with automated diagnostic software that interprets route-based measurements — reducing the specialist expertise requirement while still providing the fault detection capability that vibration analysis delivers.
Every bearing in a manufacturing facility has a story to tell in its vibration signature — weeks before it tells that story in the form of an unplanned production stop. Vibration analysis is the practice of listening to that story early enough to write a better ending.
Login
Blog
