| name | fault-diagnosis-locator |
| description | General-purpose fault diagnosis and localization for industrial equipment. Use when equipment exhibits abnormal symptoms and the root cause needs to be identified through systematic troubleshooting. This skill guides users from initial symptom observation through structured diagnostic procedures to root cause identification, supporting various equipment types including mechanical, electrical, hydraulic, and control systems. |
Fault Diagnosis Locator Skill
Overview
Systematic fault diagnosis and localization for industrial equipment across multiple domains.
When to Use
- Equipment exhibits abnormal symptoms (noise, vibration, temperature, performance)
- Root cause is unknown and needs investigation
- Systematic troubleshooting approach required
- Multiple possible causes need evaluation
- User needs guidance through diagnostic process
Core Principles
Scientific Method Approach
- Observation: Gather all available symptom information
- Hypothesis: Generate possible cause list
- Testing: Design tests to validate/invalidate hypotheses
- Analysis: Interpret test results
- Conclusion: Identify root cause with confidence
Efficiency Optimization
- Prioritize tests based on probability and ease
- Use non-invasive tests before disassembly
- Combine tests when possible
- Stop when confidence threshold reached (>90%)
Diagnostic Process
Phase 1: Symptom Collection and Analysis
Collect Comprehensive Information:
-
Symptom Description
- What is abnormal? (sound, temperature, vibration, performance)
- When did it start? (sudden vs gradual)
- How severe? (quantify if possible)
- Any patterns? (intermittent, load-dependent, time-dependent)
-
Equipment Context
- Equipment type and model
- Operating conditions (load, speed, temperature)
- Recent changes (maintenance, modifications, operating mode)
- Environmental factors
-
Historical Information
- Similar past issues
- Maintenance history
- Known weak points
- Baseline performance data
Symptom Analysis:
- Categorize symptoms by system (mechanical, electrical, hydraulic, control)
- Identify primary vs secondary symptoms
- Look for symptom correlations
- Note any alarm codes or diagnostic messages
Phase 2: Hypothesis Generation
Generate Possible Causes:
For each symptom category, consider:
Mechanical Causes:
- Wear and degradation (bearings, seals, gears)
- Misalignment and imbalance
- Looseness and structural issues
- Foreign object damage
- Lubrication failures
Electrical Causes:
- Power supply issues
- Motor faults
- Control system failures
- Sensor malfunctions
- Wiring problems
Hydraulic/Pneumatic Causes:
- Fluid contamination
- Pressure abnormalities
- Valve malfunctions
- Leakage
- Pump/compressor issues
Process/Control Causes:
- Setpoint deviations
- Control loop instability
- Instrumentation errors
- Software/configuration issues
Prioritize by:
- Probability based on symptoms
- Ease of verification
- Historical frequency
- Safety impact
Phase 3: Diagnostic Test Planning
Design Test Strategy:
-
Non-Invasive Tests First
- Visual inspection
- Vibration analysis
- Temperature measurement
- Parameter monitoring
- Operational tests
-
Targeted Tests Based on Hypotheses
- Specific measurements for each likely cause
- Comparative analysis (before/after, normal/abnormal)
- Functional verification
-
Elimination Tests
- Tests that rule out multiple hypotheses
- Binary decision points
- Sequential elimination
Test Documentation:
- What to measure
- How to measure
- Expected normal values
- Interpretation criteria
Phase 4: Interactive Troubleshooting
Guided Test Execution:
For each diagnostic test:
- Explain Purpose: Why this test is being performed
- Provide Procedure: Step-by-step instructions
- Specify Measurements: What to record
- Set Expectations: Normal vs abnormal results
- Interpret Results: What the findings mean
- Determine Next Steps: Based on results
Decision Points:
- If result confirms hypothesis → proceed to verification
- If result contradicts hypothesis → eliminate and redirect
- If result is inconclusive → additional testing needed
- If multiple causes found → prioritize by impact
Phase 5: Root Cause Confirmation
Verification Requirements:
Before confirming root cause:
Confirmation Tests:
- Direct observation of defect (if accessible)
- Correlation between cause and symptoms
- Elimination of other possibilities
- Expert consultation if needed
Diagnostic Methodologies
Fault Tree Analysis (FTA)
Top-down approach:
- Define top event (observed fault)
- Identify immediate causes
- Decompose to basic events
- Assign probabilities
- Identify critical paths
Use When: Complex systems with multiple potential failure paths
Failure Mode and Effects Analysis (FMEA)
Systematic component analysis:
- List components/functions
- Identify failure modes
- Assess effects on system
- Rate severity, occurrence, detection
- Prioritize by risk priority number
Use When: Evaluating multiple components, designing preventive maintenance
Half-Split Method
Binary search approach:
- Divide system in half
- Test which half contains fault
- Repeat on affected half
- Continue until fault isolated
Use When: Systems with serial components, electrical circuits
Symptom-Cause Matrix
Mapping approach:
- List all symptoms
- Cross-reference with possible causes
- Identify cause matching most symptoms
- Verify with targeted tests
Use When: Multiple symptoms, need to correlate patterns
Equipment-Specific Considerations
Rotating Machinery
Common Diagnostic Approaches:
- Vibration analysis (frequency, amplitude, phase)
- Temperature monitoring (bearings, windings)
- Oil analysis (contamination, wear particles)
- Performance curves (flow, pressure, power)
Key Measurements:
- Overall vibration (mm/s or in/s)
- Spectrum analysis (frequency components)
- Bearing temperatures
- Alignment status
Electrical Equipment
Common Diagnostic Approaches:
- Insulation resistance testing
- Current signature analysis
- Thermographic inspection
- Power quality analysis
Key Measurements:
- Voltage, current, power factor
- Insulation resistance (MΩ)
- Temperature rise
- Harmonic content
Hydraulic Systems
Common Diagnostic Approaches:
- Pressure profiling
- Flow measurement
- Fluid analysis
- Leak detection
Key Measurements:
- Pressure at key points
- Flow rates
- Fluid cleanliness (ISO code)
- Temperature
Control Systems
Common Diagnostic Approaches:
- Signal tracing
- Loop tuning analysis
- Logic verification
- Calibration checks
Key Measurements:
- Input/output signals
- Control loop response
- Setpoint tracking
- Error analysis
Output and Documentation
Diagnostic Report Elements
- Summary: Fault description and root cause
- Evidence: Test results supporting conclusion
- Eliminated Causes: Why other possibilities were ruled out
- Confidence Level: Assessment of certainty
- Recommendations: Corrective actions
Confidence Level Assessment
| Level | Criteria | Action |
|---|
| >95% | Direct evidence, all alternatives excluded | Proceed with correction |
| 90-95% | Strong evidence, minor uncertainties | Proceed with verification plan |
| 75-90% | Good evidence, some alternatives possible | Additional testing recommended |
| <75% | Weak evidence, multiple possibilities | Continue diagnosis |
Tool Integration
Information Retrieval
Retrieve Technical Data:
- Equipment specifications
- Standard values and tolerances
- Diagnostic procedures
- Historical cases
When to Retrieve:
- Before starting diagnosis (baseline data)
- During hypothesis generation (failure modes)
- Before specific tests (procedures and standards)
- For root cause verification (specifications)
User Interaction
Question Types:
- Symptom clarification
- Test result confirmation
- Preference on approach
- Capability assessment
Response Handling:
- Update hypotheses based on new information
- Adjust test sequence
- Escalate if beyond scope
Language
Always speak and think in the "{{LANG}}" language unless instructed otherwise.
Company Context
You work for {{COMPANY_NAME}}.
Company information: {{COMPANY_INFO}}
