up to 85k€
ESTIMATED AVOIDED COST bearing replacement and downtime.
Industry
Asset
Problem
Avoided
FAN BEARING CLEARENCE
The diagnosis was generated without additional sensors, relying solely on electrical measurements.
During normal steady-state operation, PreditMot identified a sudden and sustained upward trend in the Mechanical Severity Indicator, exceeding predefined alarm thresholds.
KEY OBSERVATIONS
– The deviation occurred without changes in process demand
– The increase was abrupt and persistent
– Alarm severity escalated rapidly, triggering a high-criticality alert
This behavior indicated abnormal mechanical looseness inconsistent with normal fan operation.
A detailed frequency-domain analysis of the stator current revealed:
- Emerging low-frequency modulation components associated with mechanical looseness
- Amplification of harmonics related to torque oscillations
- Absence (or very low magnitude) of these components under baseline conditions
The spectral pattern was consistent with progressive bearing clearance affecting rotor stability and load transmission.
CLIENT ACTION
Upon receiving the alert, the client decided to proactively stop the equipment for inspection.
A targeted mechanical inspection of the fan assembly was carried out and confirms the existence of:
– Excessive clearance in the bearing housings
After corrective action, the indicator returned to nominal values, confirming direct correlation between the electrical signature and the mechanical condition.
IMPACT & BENEFITS
✅ Prevention of catastrophic fan failure
✅ Avoidance of unplanned maintenance downtime
✅ Reduced corrective maintenance scope
✅ Increased asset reliability
KEY TECHNICAL TAKEAWAY
This case demonstrates how electrical variables can reliably reflect mechanical degradation and how ESA enables early, physics-based fault detection in industrial assets, even in harsh industrial environments where conventional sensors are difficult to deploy.
KEY TECHNICAL TAKEAWAY
Electrical Signature Analysis can identify stator winding insulation degradation at an early stage by detecting subtle changes in the motor’s electrical behaviour.
This case demonstrates how PreditMot enables physics-based fault detection using existing electrical signals, providing actionable insights without requiring intrusive sensors or machine modifications.
Access the Complete Case Study
Download the full case study or book a demo.
Want to know more?
Talk to an expert today!
