Digital Partial Discharge Tester: Power Cable Diagnostics, Location Techniques, and Aging Assessment
Power cables represent a substantial investment for utilities and industrial facilities. Insulation degradation—often caused by water trees, voids, or mechanical damage—progresses silently until failure occurs. A digital partial discharge tester enables early detection of these defects, but successful cable diagnostics requires specialized techniques for PD location and aging assessment. This article covers cable-specific PD measurement methods, location algorithms, and interpretation guidelines for extruded and paper-insulated cables.
Cable PD Fundamentals: Why Location Matters
Unlike transformers or rotating machines, power cables can extend for kilometers. Knowing that PD exists is insufficient; you must know where. A digital partial discharge tester for cable applications must include location capability. PD pulses travel along the cable at known velocity (typically 160–170 m/μs for XLPE, 150–160 m/μs for PILC). By measuring the time difference between direct and reflected pulses (or between two sensors), the tester calculates distance to the defect with typical accuracy of 1–3% of cable length.
Offline Cable PD Testing (Damped AC and VLF Methods)
For de-energized cables, two main energization methods are used with a digital partial discharge tester:
| Method | Typical Frequency | Voltage Range | Best For |
|---|---|---|---|
| Damped AC (DAC) | 20–500 Hz (decaying oscillation) | Up to 200 kV | XLPE cables >10 km, joint and termination testing |
| Very Low Frequency (VLF) | 0.01–0.1 Hz (sinusoidal) | Up to 100 kV | Medium voltage cables (5–35 kV), portable field use |
| Resonant AC | 20–300 Hz (continuous) | Up to 500 kV | Factory acceptance, long cable HVAC systems |
| Power frequency (50/60 Hz) | 50 or 60 Hz | Limited by available test set | Short cables, laboratory studies |
PD Location Techniques for Cables
A digital partial discharge tester implements one or both of these location methods:
Time Domain Reflectometry (TDR) Method
Requires a single sensor at one cable end. The tester measures the time (Δt) between the direct PD pulse (traveling directly from defect to sensor) and the reflected pulse (traveling from defect to far end and back to sensor). Distance from sensor = (v × Δt) / 2, where v is the velocity of propagation. This method works well for cables up to 5 km and defect distances beyond 50 m from the sensor. Accuracy degrades for near-end defects because direct and reflected pulses overlap.
Dual-Sensor Method (Time Difference of Arrival)
Uses sensors at both cable ends (or two locations along accessible cable route). The digital partial discharge tester calculates the time difference (Δt) between the same PD pulse arriving at each sensor. Distance from sensor A = (L + v × Δt) / 2, where L is total cable length. This method works for any defect location but requires access to both ends—not always possible for long transmission cables or cables with splices blocking signal propagation.
Interpreting Cable PD Patterns by Defect Type
The digital partial discharge tester's PRPD pattern reveals the nature of the defect:
Defective termination (poor workmanship): High magnitude PD (500–5000 pC), occurring at both positive and negative voltage peaks. Pattern often asymmetrical (stronger on one polarity). Location typically within 10 meters of termination end. Requires immediate attention—risk of termination flashover.
Damaged joint or splice: Moderate PD (200–1000 pC), symmetrical pattern around voltage peaks. Location matches known joint position. Often caused by moisture ingress or installation error. Plan replacement within 6–12 months.
Water treeing in XLPE (extruded cables): Low level PD (20–200 pC) appearing at 45°–80° and 225°–260° phase angles. Pattern evolves over years. Location distributed along cable, not a single point. Indicates general aging rather than discrete defect. Monitor annually.
Void in insulation (manufacturing defect): PD appears internally, with stable magnitude and phase position. Often detectable at lower voltages (e.g., 50% of rated). Location varies. For new cable installations, any void-type PD above 50 pC warrants investigation.
Semicon layer protrusion (stress enhancement): Very sharp PD pulses with extremely fast rise time (<5 ns). Pattern shows pulses near voltage zero crossing. Typically found during factory testing. Rejection threshold depends on cable voltage class.
Online PD Testing for Cables (Energized)
Online cable PD testing uses permanently installed or portable HFCT sensors clamped around grounding leads or directly on the cable shield. Advantages: no outage required, captures PD under actual load and temperature. Disadvantages: cannot be calibrated in pC (measure in mV), interference from nearby cables and substation noise. A digital partial discharge tester with multi-channel and differential measurement capability is essential for online cable testing.
For online location, consider the following:
Standard TDR location fails because PD pulses are not reflected from the far end when the cable is energized.
Use multi-sensor correlation along the cable route (e.g., at manholes every 200–500 meters). The tester triangulates the defect by comparing arrival times at 3+ sensors.
Alternatively, use a gating technique: apply a high-frequency pulse from a coupler and measure its propagation; defects create reflections detectable even with cable energized.
Aging Assessment: PD Fingerprinting and Trending
For cable fleet management, a digital partial discharge tester should store historical measurements per cable segment. Key trending parameters:
| Parameter | Meaning | Alarm Threshold (annual increase) |
|---|---|---|
| Peak PD magnitude (pC) | Worst-case discharge intensity | >20% / year |
| Average PD magnitude (pC) | General activity level | >10% / year |
| PD pulse repetition rate | Frequency of damaging events | Doubling in 12 months |
| Inception voltage (kV) | Voltage at which PD starts | Decrease >15% |
| Phase width of PD activity | Spread of discharge across AC cycle | Increase >30 degrees |
Case Study: Locating a Failing Cable Joint
A 15 kV XLPE feeder cable experienced two outages in six months, both cleared by reclosing. A digital partial discharge tester was connected using VLF energization at 0.1 Hz, 19 kV peak (rated voltage). The tester measured PD at 850 pC, with pattern suggesting a joint defect. Using dual-sensor location (sensors at substation and remote end, 2.3 km apart), the tester calculated distance at 1,487 meters from the substation. Crews excavated at that location and found a factory joint with punctured insulation tape. Replacement joint eliminated PD, and the cable operated without further incidents for 5+ years.
Challenges Specific to Cable PD Testing
Attenuation: High-frequency PD pulses attenuate rapidly in cables (typically 10–30 dB/km). For cables longer than 5 km, locate sensors every 2–3 km or use lower frequency band PD detection (100–500 kHz).
Dispersion: Different frequency components of a PD pulse travel at different speeds, broadening the pulse and complicating location. A digital partial discharge tester with cross-correlation algorithms compensates for dispersion.
Reflections from splices: Each splice creates partial reflections, generating multiple phantom PD indications. Use a cable map with known splice positions to validate calculated locations.
Adjacent cable interference: In multi-circuit trenches, PD from one cable can couple inductively to neighboring cables. Always test with all adjacent cables de-energized when possible.
Selecting a Digital Partial Discharge Tester for Cable Work
Prioritize these features for cable-focused applications:
Built-in velocity calibration using known cable reference (e.g., a pulse injected at the far end).
Support for DAC, VLF, and power frequency energization (or compatibility with separate VLF/DAC sources via external trigger).
Location accuracy specification of <2% of cable length (or <50 meters, whichever larger).
Ability to store cable route maps with joint locations for automated correlation.
Cloud or network database for fleet-wide cable asset management.
IP54 or higher rating for field use in manholes and outdoor substations.
Recommended Test Sequence for Cable Acceptance or Maintenance
Perform insulation resistance test (megger) to verify no gross breakdown.
Connect digital partial discharge tester and energize cable to 0.5× rated voltage. Measure background noise.
Increase voltage in steps of 0.1× rated up to 1.5× rated for acceptance (or 1.0× rated for maintenance). Hold each step for 2–5 minutes.
Record PD inception voltage (PDIV). For acceptance, PDIV must exceed 1.2× rated voltage for extruded cables per IEEE 400.
If PD detected above acceptable threshold, perform location measurement and mark defect position.
Repeat test after repair to confirm PD reduction to <50 pC at rated voltage.
A digital partial discharge tester tailored for power cable applications is an indispensable tool for utilities, industrial plants, and cable manufacturers. The combination of sensitive PD detection, accurate location algorithms, and pattern recognition enables targeted repairs, avoiding unnecessary cable replacement. When integrated into a condition-based maintenance program, cable PD testing reduces outage costs by 50–80% compared to reactive failure-based approaches.
