Bearing Fault Diagnosis

Bearing Fault Diagnosis

Evidence-based bearing fault detection using vibration signals. Orchestrate MCP tools in a precise diagnostic sequence with decision gates at each step.

Prerequisite: The predictive-maintenance-mcp MCP server must be connected.

Workflow

Step 1 — Signal Discovery

Call list_stored_signals() to check cached signals, or list_signals() to browse the data/ directory. If no signal is loaded, ask the user for a file path and call load_signal(file_path=..., signal_id=...). Ask which signal to analyze if multiple are available.

Step 2 — Statistical Screening

Call extract_features_from_signal(signal_id=...).

Evaluate screening flags (excess kurtosis, Fisher convention):

Indicator	Threshold	Meaning
Kurtosis > 0	Mild	Non-Gaussian content (possible impulses)
Kurtosis > 3	Moderate	Significant impulsive content
Kurtosis > 6	Severe	Likely bearing damage
Crest Factor > 4	Mild	Impulsiveness present
Crest Factor > 6	Strong	Strong impulsiveness

Decision gate: If Kurtosis < 0 AND Crest Factor < 3 the signal shows no impulsive content. Report "No bearing fault indicators in time-domain screening" and ask if the user wants to continue anyway.

Step 3 — Spectral Analysis (FFT)

Call analyze_fft(signal_id=...).

Identify:

• Dominant frequencies and their harmonics
• Shaft frequency (1x RPM) and multiples
• Broadband energy increase

If operating speed is unknown, ask the user for RPM or shaft frequency.

Step 4 — Bearing Characteristic Frequencies

If bearing designation is known, call lookup_bearing_and_compute_tool(bearing_query=..., shaft_rpm=...) to look up the bearing in the catalog AND compute fault frequencies in one step.

Otherwise call calculate_bearing_characteristic_frequencies(...) with manual geometry (n_balls, d_ball, d_pitch, contact_angle, shaft_rpm).

If the bearing designation is unknown, ask the user or try extract_manual_specs(manual_name=...) to pull it from a machine manual.

Expected frequencies:

• BPFO — Ball Pass Frequency Outer race
• BPFI — Ball Pass Frequency Inner race
• BSF — Ball Spin Frequency
• FTF — Fundamental Train (cage) Frequency

Step 5 — Envelope Analysis

Call analyze_envelope(signal_id=..., filter_low=..., filter_high=...).

Adjust the filter band based on Step 3:

• If a resonance was found in FFT, center the band around it
• Default band: 500–5000 Hz for general bearing analysis
• High-speed machines: shift band higher

Call plot_envelope(...) for a visual report.

Step 6 — Evidence Matching

Compare envelope peaks against characteristic frequencies using check_bearing_faults_direct(signal_id=..., shaft_rpm=..., ...).

Peak matches	Harmonics	+ High Kurtosis	Diagnosis
BPFO +/- 2%	2x, 3x present	Yes	Possible outer race fault
BPFI +/- 2%	+ sidebands at shaft freq	Yes	Possible inner race fault
BSF +/- 2%	2x present	Yes	Possible ball defect
FTF +/- 2%	Irregular spacing	Moderate	Possible cage fault
No matches	—	—	Inconclusive

Confidence levels:

• Confirmed: Peak + harmonics + high kurtosis + additional indicator
• Possible/consistent with: Peak present but missing corroboration
• Inconclusive: Insufficient evidence

Step 7 — ISO 20816-3 Severity

Call evaluate_iso_20816(signal_id=..., machine_group=2, support_type="rigid").

Report vibration zone (A/B/C/D) and urgency level. Confirm signal units with the user before calling — wrong units invalidate the result.

Step 8 — Report Generation

Call generate_envelope_report(...) and generate_fft_report(...). Optionally call generate_diagnostic_report_docx(...) for a full DOCX report. Inform the user of report file locations.

Troubleshooting

No peaks in envelope spectrum — Wrong filter band or signal too short. Try a wider band (200–10000 Hz) or analyze the full signal without segmentation.

Unknown bearing type — Ask for bearing designation (e.g., SKF 6205) and call search_bearing_catalog(). If unavailable, check the machine manual with extract_manual_specs().

Kurtosis high but no envelope peaks — Impulses may come from a non-bearing source (gear mesh, electrical). Suggest using the gear-diagnosis skill or check for electrical frequencies (line frequency harmonics).

Important Notes

• Always use cautious diagnostic language: "consistent with", "possible fault"
• Never claim a definitive fault without multiple corroborating indicators
• Confirm signal units before ISO severity assessment
• All processing happens locally — raw signal data never leaves the machine