Every rotating machine has a heartbeat. A healthy motor, pump, or fan hums along with a steady, repeatable vibration, and when something starts to go wrong, that heartbeat changes. Imbalance, misalignment, looseness, and bearing wear each leave their own pattern long before the machine gets loud, hot, or stops. Vibration analysis is how you read those patterns, and it is widely treated as the backbone of predictive maintenance for rotating equipment. But a reading only pays off when it triggers action: the right work order, on the right asset, before the failure. OxMaint AI maintenance management software keeps each machine's measurement points, history, and alarms in one place and turns a vibration finding into a planned repair.
Vibration Analysis for Predictive Maintenance: The Complete Guide
Learn to read what your machines are telling you: which fault shows up where in the spectrum, how serious it is, and what to do about it. OxMaint AI links every finding to the asset, the alarm, and the work order.
imbalance, misalignment, looseness, bearing wear: the bulk of rotating-machine problems
3 axes
horizontal, vertical, and axial at every bearing
Zones A–D
ISO 20816 severity bands, from good to damaging
4 stages
a typical bearing degradation path
The Fault Finder: Four Patterns to Learn First
You don't need to memorize hundreds of rare faults. Most problems on rotating machines come down to four, and each one has a recognizable spectrum shape. Compare a reading against these patterns, and always against the machine's own healthy baseline. Start free and record the diagnosis on each asset.
Imbalance
PatternDominant peak at 1X running speed
DirectionStrongest radial, not axial
Common causeDirty or damaged blades, lost balance weight
FixClean, inspect, and balance the rotor
Misalignment
PatternElevated 1X and 2X, sometimes 3X
DirectionHigh axial, or 2X radial for parallel offset
Common causePoor coupling alignment, soft foot, thermal growth
FixLaser-align the shafts and check the foundation
Mechanical Looseness
PatternA run of harmonics of 1X, sometimes with half-orders
DirectionOften strongest in one direction (vertical)
Common causeLoose bolts, cracked base, worn bearing fit
FixTighten, shim, and repair the base or housing
Bearing Wear
PatternNon-synchronous peaks at bearing defect frequencies
DirectionRadial, best seen with envelope analysis
Common causePoor lubrication, contamination, overload
FixCheck lubricant, then plan bearing replacement
Diagnosis Is Only Half the Job. The Repair Has to Happen.
When a reading crosses your alarm level, OxMaint AI raises a work order on that exact machine with the fault, the measurement point, and the recommended action attached. Parts and a planned window follow, so the bearing is changed on your schedule, not the machine's.
One RMS number to compare against ISO limits and your own baseline.
What is wrong?
FFT Spectrum
Splits vibration by frequency so each fault shows its own peaks.
Is something impacting?
Time Waveform
Shows impacts, rubs, and looseness that a spectrum can hide.
Is a bearing starting to fail?
Envelope Analysis
Pulls faint bearing defect pulses out of high-frequency noise.
Imbalance or misalignment?
Phase Analysis
Compares phase across a coupling or point to the heavy spot.
How Serious Is It? ISO 20816 Severity Zones
ISO 20816 (which replaced ISO 10816) groups overall vibration into four zones, so everyone reads severity the same way. Smaller machines often fall outside the standard, so rely on supplier limits and your own baseline trend for those. Start free and map each asset to a zone limit.
A
Good
Typical of newly commissioned machines.
B
Acceptable
Fine for long-term operation. Keep trending.
C
Restricted
Not suited to long-term running. Plan the fix.
D
Damaging
Severe enough to cause damage. Act now.
Remember: a single high number doesn't name the fault. The same overall level can come from imbalance, misalignment, looseness, or a failing bearing, and often a mix. Always check the spectrum.
The Bearing Failure Path: Four Stages
Bearings usually degrade in a recognizable sequence, and vibration sees the early stages long before heat or noise do. The earlier you catch it, the cheaper the fix. Book a demo to track bearing health per asset.
Stage 1
Earliest signs
Faint high-frequency changes. Only sensitive methods see it.
Stage 2
Defect frequencies appear
Distinct peaks emerge. Time to plan.
Stage 3
Harmonics & sidebands
Damage spreads. Schedule the replacement soon.
Stage 4
Noise floor rises
Heat and noise climb. Failure is close.
Cheapest to fixMost costly
Route Checklist: Measure the Same Way Every Time
Reliable trending depends on repeatable measurement. Take readings at every bearing in three directions, at the same points and the same operating condition. Tick off a route like this one. Start free and turn it into a mobile route.
Each machine with its measurement points, baselines, and full repair history.
📱
Mobile Route Checklists
Technicians follow the route, log readings, and add photos from the floor.
🚨
Alarm-to-Work-Order
A crossed limit opens a work order with fault details and priority.
📈
Trends & Predictive Planning
Track asset health over time and schedule repairs before the failure window.
🔌
Sensor & PLC Integration
Bring in condition data so thresholds are watched in one place.
🧰
Parts & Planning
Line up bearings, seals, and a maintenance window ahead of time.
Frequently Asked Questions
What is vibration analysis in predictive maintenance?
It is the practice of measuring and interpreting machine vibration to detect faults such as imbalance, misalignment, looseness, and bearing wear early, so repairs can be planned before a failure. Start free and manage it in OxMaint AI.
What faults can vibration analysis detect?
Mainly imbalance, shaft misalignment, mechanical looseness, and bearing defects, plus gear mesh problems and resonance. These show up as different patterns in the frequency spectrum. Book a demo to see fault tracking.
How do I tell imbalance from misalignment?
Imbalance gives a dominant 1X peak that is strongest radially. Misalignment raises 1X and 2X, with high axial vibration or strong 2X radially. Phase readings across the coupling help confirm. Sign up free to record your findings.
How often should I take vibration readings?
It depends on criticality. Many plants run monthly routes on general machines and use continuous sensors for critical or fast-failing assets. Shorten the interval when a trend starts to climb. Book a demo to schedule routes.
What are the ISO 20816 zones?
Zone A is good, B is acceptable for long-term running, C is restricted and needs a planned fix, and D is vibration severe enough to cause damage. Smaller machines may fall outside the standard, so use supplier limits and a baseline. Start free and set limits per asset.
How does a CMMS support vibration analysis?
It stores each machine's points, baselines, and history, schedules routes, raises a work order when a limit is crossed, and tracks the repair to completion. Book a demo of the workflow.
Hear the Failure Before It Happens.
Your machines are already telling you where they hurt. Pair vibration analysis with OxMaint AI and every warning becomes a work order, a part on the shelf, and a repair done on your terms.
.png)
