Cases/NVH/C03
NVHWind / chemical / steel predictive-maintenance providers

Envelope-Demodulation Bearing Fault Diagnosis

01

Pain Points

  • !Early faults are buried in broadband vibration; the raw spectrum shows no inner/outer race, rolling-element, or cage signature.
  • !The four characteristic frequencies must be computed on the spot from geometry and RPM — error-prone by hand.
  • !"Outer race of fan #3 has failed" lacks a reproducible diagnostic path.
02

The Tinia Approach

High-frequency band-pass + Hilbert envelope + characteristic-frequency matching pinpoints the fault location and severity trend — reuse on a new model by changing only the geometry parameters.

Resonance-band band-passEnvelope demodulationEnvelope-spectrum FFTCharacteristic-frequency matchingLabel verification
03

Nodes Used

IIR Filteriir_filter
Resonance-band band-pass preprocessing
Envelope Demodulationenvelope_demod
Extract the modulation envelope via Hilbert transform
FFT Spectrum Analysisfft_spectrum
Compute the envelope spectrum
Indicator Mathindicator_math
Compute and match BPFO/BPFI/BSF/FTF frequencies
Spectrum Viewerspectrum_viewer
Label characteristic-frequency positions and verify harmonics

Core nodes; add or remove steps as needed for your data and standards — the node graph is always editable.

04

Expected Output

  • Envelope spectrum + four characteristic-frequency markers + harmonic verification.
  • Fault-location identification (outer race / inner race / rolling element / cage).
  • Severity trend over time to support maintenance scheduling.
Reference Standards / Engineering PracticeAlgorithm standardization (no single mandatory standard)

Want to run this workflow on your own data?

Request a Trial
NVH

Other cases in this domain