Vibration Diagnostics of Rolling-Element Bearings: How to Predict a Failure Before It Happens
Rolling-element bearings are a critically important part of any rotating machine. Statistics show that up to 80% of bearing failures can be predicted several weeks or months before complete destruction by using vibration diagnostics.
This article is an advanced guide to diagnosing bearing faults, complementing the introductory article on vibration diagnostics. Here we examine in detail how a faulty bearing "sounds" in the vibration spectrum and how to tell apart faults in the various bearing components.
Article level: 🔴 Advanced
If you are new to the subject, we recommend reading the introductory guide to vibration diagnostics first
The physics of bearing faults
When a fault (a pit, a spall or a crack) appears on a raceway surface (a ring or a rolling element), a short impact pulse is generated each time a rolling element rolls over that fault.
These pulses repeat at a particular frequency, characteristic of each bearing component, which depends on:
- The geometry of the bearing (number of rolling elements, diameters)
- The rotational speed of the shaft
It is precisely these non-synchronous frequencies (not multiples of the rotational speed!) that are the "signature" of bearing faults in the vibration spectrum.
Characteristic bearing fault frequencies
Each bearing component has its own characteristic fault frequency:
| Frequency | Meaning | Component |
|---|---|---|
| BPFO | Ball Pass Frequency Outer race | The frequency at which rolling elements pass over a fault on the outer race |
| BPFI | Ball Pass Frequency Inner race | The frequency at which rolling elements pass over a fault on the inner race |
| BSF | Ball Spin Frequency | The frequency at which a rolling element spins about its own axis |
| FTF | Fundamental Train Frequency | The rotational frequency of the cage |
Professional bearing vibration diagnostics
We assess the condition of bearings and forecast when replacement will be needed
Request diagnosticsOuter-race fault (BPFO – Ball Pass Frequency Outer race)
Physical description: a fault (a pit, a spall or a crack) on the raceway of the bearing's outer ring. Each time a rolling element rolls over this damage, an impact pulse is generated.
Spectrum description: the vibration spectrum shows a series of peaks corresponding to the outer-race fault frequency and its harmonics. These peaks usually lie at higher frequencies (they are not whole multiples of the shaft rotational frequency) and mark each moment a rolling element passes over the fault.
Inner-race fault (BPFI – Ball Pass Frequency Inner race)
Spectrum description: with an inner-race fault, the spectrum shows several pronounced peaks at the inner-race fault frequency and its harmonics. In addition, each of these fault-frequency peaks is usually accompanied by sidebands spaced at the rotational frequency (1×).
Rolling-element fault (BSF – Ball Spin Frequency)
Spectrum description: a fault on a rolling element (a ball or a roller) produces vibration at the spin frequency of the rolling element and its harmonics. The spectrum shows a series of peaks that are not whole multiples of the shaft rotational frequency but multiples of the ball/roller spin frequency (BSF).
Cage fault (FTF – Fundamental Train Frequency)
Spectrum description: a cage fault in a rolling-element bearing produces vibration at the rotational frequency of the cage — the Fundamental Train Frequency (FTF) — and its harmonics. These frequencies are usually sub-synchronous (lower than the shaft rotational frequency).
Stages of bearing fault development
A bearing fault develops in stages:
- Initial stage: the first faint peak appears at the characteristic frequency
- Development: the peak amplitude grows and harmonics appear
- Progressive: numerous harmonics, sidebands, and a rising overall "noise" level in the high-frequency region
- Critical: very high peaks, broadband noise, unstable readings
Monitoring recommendations:
- Once bearing frequencies appear — step up monitoring (measure more often)
- Check the lubrication
- Start planning to replace the bearing at the next opportunity
- A sharp rise in amplitude is the signal for urgent replacement
The advantage of vibration diagnostics: detecting bearing faults 2–6 months before complete destruction gives ample time to plan the repair, order spare parts and choose the best moment to shut the machine down.
Conclusion
Diagnosing rolling-element bearings from the vibration spectrum is a powerful predictive-maintenance tool. Understanding the characteristic frequencies (BPFO, BPFI, BSF, FTF) and being able to recognise them allows you to:
- Detect faults at an early stage
- Plan replacements well in advance
- Avoid emergency shutdowns
- Reduce repair costs
Modern vibration analysers such as the Balanset-1A make it possible to obtain detailed spectra and to detect bearing frequencies even at the early stage of fault development.
Bearing diagnostics
Instruments for diagnostics and professional vibration diagnostics services
Balanset-1A instrument
A vibration analyser for bearing diagnostics (BPFO, BPFI, BSF, FTF)
Buy the instrument