1. Introduction

Industrial electric motors are the workhorses of modern manufacturing, driving pumps, fans, conveyors, compressors, and countless other machines. Their unscheduled downtime can result in catastrophic production losses, costly repairs, and significant safety hazards. Consequently, proactive and predictive maintenance (PdM) strategies have become paramount for ensuring motor reliability and optimizing operational efficiency. At the core of many PdM programs lies the continuous or periodic monitoring of motor electrical signatures.

While traditional solid-core current transformers (CTs) offer high accuracy, their requirement for installation on de-energized and disconnected cables makes them impractical for many monitoring applications. Split-core current transformers have emerged as the enabling technology for non-intrusive motor monitoring. This case study explores the application of split-core CTs within the predictive maintenance program of a large chemical processing plant, detailing the challenges, implementation process, results, and best practices derived from this real-world scenario.

2. Background: The Challenge at the Chemical Plant

The "ChemSite" facility (a pseudonym) operates 24/7, with over 500 critical motors ranging from 5 HP to 2500 HP. The plant experienced an average of three to five unexpected motor failures per year, primarily due to:

• •

  Bearing Wear: Leading to increased friction and load.

• •

  Stator Winding Insulation Degradation: Often from overheating or voltage spikes.

• •

  Rotor Bar Failures: Causing torque pulsations and efficiency loss.

• •

  Load Changes: Upstream process issues leading to motor overload.

The maintenance team relied on a time-based preventive maintenance (PM) schedule for motor inspection and lubrication. However, this approach failed to detect issues that arose between PM intervals. Vibration analysis was used but required skilled technicians to collect data manually, leaving gaps in coverage. The plant management mandated a shift towards a more data-driven, predictive maintenance approach to reduce unplanned downtime and maintenance costs.

3. The Solution: Implementing a Motor Monitoring System with Split-Core CTs

A project was initiated to install a continuous online motor monitoring system on 150 of the most critical motors. The key component of this system was the split-core current transducer.

3.1. Why Split-Core CTs?

The decision to use split-core CTs was driven by several critical factors:

• •

  Non-Intrusive Installation: The number one advantage. The hinged core could be clamped around existing motor feeder cables without any need to shut down equipment, disconnect wires, or alter the electrical installation. This was essential for a live production plant.

• •

  Safety: Eliminating the need to work on live, exposed electrical terminations significantly reduced the risk for installation technicians.

• •

  Cost-Effectiveness: The savings from avoiding production downtime for installation far outweighed the slightly higher unit cost of a split-core CT compared to a solid-core one.

• •

  Flexibility and Retrofit: They could be easily added to existing motors at any time, making them perfect for a phased PdM rollout.

3.2. System Design and Component Selection

The monitoring system consisted of:

• •

  Split-Core Current Transformers: Class 1.0 accuracy was selected as it provided sufficient precision for trend-based predictive analysis, though not for revenue-grade metering. A 5A secondary output was chosen for compatibility with the monitoring hardware. The CTs were rated for the full load current of each motor with an additional margin.

• •

  Data Acquisition Nodes: These units, installed locally in motor control centers (MCCs), converted the analog CT signals into digital data. They also included inputs for voltage and temperature sensors.

• •

  Network Gateway: Data from the nodes was aggregated and transmitted via the plant's industrial Ethernet network.

• •

  Cloud-Based Analytics Software: This platform received the data, stored it in a time-series database, and performed real-time analysis, providing dashboards, alerts, and diagnostic tools.

3.3. Key Installation Considerations

Proper installation was critical for data accuracy:

• •

  CT Placement: Each CT was installed on a single phase conductor (not a bundled cable containing all three phases). The phase conductor was centered within the CT aperture to ensure optimal performance.

• •

  Orientation and Closing: The CT was closed securely to ensure a full magnetic circuit. The locking mechanism was engaged to prevent the CT from accidentally opening due to vibration.

• •

  Polarity: While less critical for basic RMS current measurement, correct orientation (P1 mark facing the source) was ensured for future potential use in power quality analysis.

• •

  Secondary Wiring: Shielded, twisted-pair cabling was used for the secondary leads from the CT to the data acquisition node to minimize electromagnetic interference (EMI) from the high-power environment. The leads were kept as short as possible, and the shield was grounded at one end only to prevent ground loops.

• •

  Burden Calculation: The input impedance of the data acquisition node was extremely low (effectively a short circuit). The burden was dominated by the wire resistance. The correct wire gauge was selected to ensure the total burden did not exceed the CT's rating, preventing saturation and measurement error.

4. The Case Study: Bearing Failure Prediction on a Critical Pump Motor

4.1. The Asset

A 400 HP, 4kV vertical turbine pump motor responsible for supplying feedstock to a primary reactor. A failure of this motor would force a reactor shutdown, costing over $250,000 per day in lost production.

4.2. The Data and The Alert

The motor was fitted with split-core CTs on all three phases. The monitoring software tracked several key parameters, including:

• •

  RMS Current (All Phases)

• •

  Current Unbalance

• •

  Percent Load

• •

  Power (kW)

• •

  The Full Current Waveform (sampled at 4 kHz for advanced analysis)

Eight months after installation, the software's algorithm triggered a "Yellow Alert" for this motor. The trend analysis indicated a gradual, consistent 8% increase in RMS current draw over a six-week period, while the process flow (and therefore the motor load) had remained constant. The current unbalance and voltage parameters were normal.

4.3. The Diagnostic Process

The maintenance team received the alert via email and reviewed the motor's dashboard. The steady increase in amperage with constant load indicated an increase in the mechanical torque required to drive the pump. This pointed to a mechanical issue, not an electrical one. The vibration analysis team was dispatched to collect high-resolution data.

The vibration spectra revealed a classic signature: high amplitudes at the fundamental bearing frequencies (BPFO - Ball Pass Frequency Outer race), confirming the early stages of a bearing failure on the pump end. The electrical trend from the split-core CTs had provided the early warning; vibration analysis provided the specific diagnosis.

4.4. The Outcome: From Reactive to Predictive

Based on this predictive warning, the maintenance team:

1. 1.

   Planned the Repair: The bearing and necessary seals and gaskets were ordered.

2. 2.

   Scheduled Downtime: The work was scheduled for the next planned maintenance window, three weeks later.

3. 3.

   Executed the Repair: The motor was shut down, the failed bearing was replaced, and the motor was returned to service. Inspection confirmed spalling on the outer race.

4. 4.

   Verified the Fix: Post-repair current readings returned to their original baseline values, confirming the resolution of the fault.

The total cost of the planned repair was approximately $15,000(parts,labor,andplannedlostproduction).Thiscomparedtoapotentialcostofover$750,000 for an unexpected failure, which would have included emergency repairs, lost production from an unscheduled shutdown, and possible collateral damage to the pump.

5. Broader Applications and Benefits Realized

Beyond bearing faults, the split-core CT-based system provided value across the plant:

• •

  Detection of Stator Winding Issues: A growing current unbalance trend on a fan motor led to the discovery of deteriorating insulation in one phase.

• •

  Rotor Bar Fault Detection: Advanced analysis of the current waveform (Current Signature Analysis) identified sideband frequencies around the line frequency, indicating cracked rotor bars in a large compressor motor.

• •

  Load Monitoring: Operators could verify that motors were operating within their optimal load range, preventing underloading (inefficiency) and overloading (overheating).

• •

  Process Monitoring: Changes in current draw often reflected process conditions, providing operators with an additional tool for troubleshooting flow or pressure issues.

The overall benefits realized included:

• •

  A 65% reduction in unplanned motor downtime in the first year.

• •

  Extended motor life through early fault detection and timely intervention.

• •

  Improved safety by reducing the need for emergency troubleshooting on live equipment.

• •

  Data-driven decision making, moving from calendar-based to condition-based maintenance tasks.

6. Limitations and Best Practices

Limitations:

• •

  Slightly Lower Accuracy: Solid-core CTs generally offer better accuracy (Class 0.5 or 0.2) due to a more perfect magnetic path.

• •

  Potential for Improper Installation: An unsecured or misaligned CT can lead to erroneous data.

• •

  Limited to AC Measurements: Cannot be used on DC circuits.

Best Practices:

1. 1.

   Select the Right CT: Ensure the rated primary current covers the motor's operating range. Choose a robust CT with a secure latching mechanism.

2. 2.

   Install with Care: Center the conductor, close the core fully, and secure it. Use appropriate wiring practices to minimize noise.

3. 3.

   Verify Installation: After installation, use a handheld clamp-meter to verify the readings from the permanent CT against a known-good portable device.

4. 4.

   Focus on Trends: The greatest power of this technology is in observing changes over time. Establish a healthy baseline for each motor.

5. 5.

   Integrate with Other Technologies: Use current monitoring as the broad, continuous "watchdog" and pair it with targeted vibration analysis, thermography, and ultrasonic testing for a comprehensive PdM program.

7. Conclusion

This case study demonstrates that split-core current transformers are far more than a simple convenience tool. They are a powerful and pragmatic enabling technology for industrial predictive maintenance. By allowing for safe, non-intrusive, and continuous monitoring of motor current, they provide a vital window into the health of both the motor and the driven equipment.

The successful implementation at the ChemSite facility underscores a critical paradigm shift: moving from reactive, failure-based maintenance to proactive, data-driven condition-based maintenance. The early detection of a developing bearing fault, leading to a planned and controlled repair, resulted in direct and substantial financial savings. The ROI was proven not just in avoided repair costs, but in the preservation of continuous production—the lifeblood of any industrial operation. For any organization relying on critical motor-driven assets, split-core CTs represent a low-risk, high-reward first step into the world of Industry 4.0 and smart, predictive maintenance.