Partial Discharge
Definition: Partial discharge is a localized electrical breakdown that crosses only part of the insulation between two conductors. The discharge stops before bridging the electrodes completely, so no short circuit forms. Each event, however, erodes the surrounding insulation material.
Key Takeaways
- Partial discharge is a localized insulation fault that does not bridge the conductors, so equipment keeps running while damage accumulates.
- The three mechanisms are internal discharge in voids, surface tracking across insulation interfaces, and corona at sharp conductor points.
- IEC 60270 governs conventional offline measurement, which reports apparent charge in picocoulombs (pC).
- Online techniques such as TEV, HFCT, ultrasonic, and UHF sensing test energized equipment without taking it out of service.
- Trending discharge activity over time gives more reliable guidance than a single reading.
What Is Partial Discharge?
Partial discharge is a localized dielectric breakdown that crosses a portion of the insulation between two conductors. The discharge bridges part of the insulating gap and stops before reaching the opposite electrode, so no fault current flows through the full insulation path and the equipment continues to operate.
That incomplete path separates it from flashover, in which insulation fails outright and a conductive channel connects the electrodes. Discharge instead occurs in three settings: in voids or delaminations inside solid insulation, along insulation surfaces, and at sharp conductor points in gas, where the effect is known as corona. Each event deposits energy into a small volume of material, and the damage compounds with repetition.
Why Partial Discharge Matters
Partial discharge (PD) is both a symptom and an accelerator of insulation degradation. The discharge signals that a defect already exists; at the same time, each event erodes the surrounding material, carbonizes pathways through the insulation, and enlarges the defect. Given enough repetitions, the eroded path eventually bridges the electrodes and the insulation fails.
The equipment affected covers much of a plant's electrical infrastructure: metal-clad switchgear, power transformers, rotating machine windings such as motor and generator stators, and cable terminations and joints. These assets typically sit at the head of the distribution system, so their condition shapes the reliability of the supply downstream.
The escalation pattern follows a consistent sequence: discharge activity appears at a defect, the defect grows as discharges repeat, insulation strength declines, and the asset eventually suffers equipment failure that can trigger unplanned downtime. Because the first stage is measurable long before the last, partial discharge testing has become a standard condition assessment technique for medium- and high-voltage equipment.
The Three PD Mechanisms
Partial discharge takes three main forms, distinguished by where the discharge occurs within the insulation system. Identifying the mechanism matters because each one progresses at a different rate and points to a different root cause.
Internal discharge takes place inside voids, delaminations, or gaps within solid insulation such as epoxy, cast resin, or oil-paper systems. These cavities form during manufacturing or develop through thermal aging and mechanical stress. Surface tracking runs across the boundary between insulation and the surrounding medium, most often on contaminated or moist bushings and terminations. Corona occurs in gas at sharp conductor points, where the local electric field concentrates enough to ionize the air.
| Mechanism | Where It Occurs | Typical Trigger |
|---|---|---|
| Internal | Voids, delaminations, and gaps inside solid insulation | Manufacturing defects, thermal aging, mechanical stress |
| Surface | Interfaces between insulation and the surrounding air, such as bushings and terminations | Moisture, contamination, pollution buildup |
| Corona | Sharp conductor points in gas | High local electric field at protrusions, damaged strands, or sharp edges |
How Partial Discharge Is Measured
The conventional measurement reports apparent charge in picocoulombs (pC). Apparent charge is the charge measurable at the equipment terminals that corresponds to the discharge inside the insulation, and it serves as the standard quantity for comparing readings across tests. The IEC 60270 standard defines the conventional offline procedure: the asset is de-energized, connected to a coupling capacitor and measuring impedance, and excited at rated or reduced voltage while discharge pulses are recorded.
Online methods measure discharge while equipment remains energized. Transient earth voltage (TEV) sensors detect the voltage transients that discharge pulses induce on the metal enclosure of switchgear. High-frequency current transformers (HFCT) clamp around cable earth connections to capture discharge current pulses. Ultrasonic and acoustic sensors pick up the sound that discharge events produce, and ultra-high-frequency (UHF) sensors capture the electromagnetic emissions from discharge inside shielded compartments such as gas-insulated switchgear.
Online testing fits naturally into a predictive maintenance program, since it gathers condition data without taking critical assets out of service.
| Method | Where Used | What It Detects |
|---|---|---|
| IEC 60270 conventional | Offline tests on transformers, rotating machines, and cables | Apparent charge in pC measured at the terminals |
| TEV | Metal-clad switchgear | Voltage transients induced on the enclosure by internal discharge |
| HFCT | Cable circuits, at earth connections or terminations | High-frequency discharge current pulses in cables and accessories |
| Ultrasonic and acoustic | Switchgear, transformers, rotating machines | Airborne ultrasound and structure-borne sound from discharge events |
| UHF | Gas-insulated switchgear, transformers, cable systems | Electromagnetic emissions from discharge inside shielded compartments |
PD Patterns and Interpretation
Raw pulse data becomes usable through phase-resolved analysis. Phase-resolved partial discharge (PRPD) patterns plot discharge magnitude against the phase angle of the AC cycle, and different defect types produce recognizable shapes: internal voids, surface tracking, and corona leave distinct signatures in the pattern.
Trending carries more weight than any individual measurement. A single reading reports the state at one moment; a series of readings shows whether activity is stable, intermittent, or growing. Rising magnitude, a higher pulse repetition rate, or discharge that appears at lower voltage indicates an accelerating defect and calls for shorter retest intervals.
Interpretation also requires separating true discharge from electrical noise. Variable-speed drives, SCR switching, radio transmitters, and loose connections generate signals that can mimic PD on some sensors. Experienced analysts cross-check with multiple sensor types, look for consistency in phase position, and confirm that pulses repeat predictably before classifying a source as discharge.
Practical Program Guidance
A partial discharge program starts with a baseline survey of medium- and high-voltage assets. The survey records each asset's initial signature, giving later tests something to compare against. Assets then move onto a periodic retest cycle, and assets showing activity get shorter intervals and closer attention.
Electrical PD testing works best alongside other inspection techniques. Infrared analysis reveals the heating that discharge and other defects produce, acoustic inspection catches audible and ultrasonic emission, and continuous asset condition monitoring fills the gap between scheduled tests. Combining methods reduces the chance that a single technique misses a developing fault.
Partial discharge testing is specialized work. Measurements on energized high-voltage equipment require qualified personnel, appropriate safety clearances, and instruments suited to the measurement type. Many plants bring in specialist contractors for offline IEC 60270 tests while running simpler online surveys with in-house teams.
Frequently Asked Questions
What is the difference between partial discharge and corona?
Corona is one of the three partial discharge mechanisms. It occurs in gas at sharp conductor points where the local electric field is strong enough to ionize the air. Internal discharge inside voids and surface tracking across insulation interfaces are the other two mechanisms.
Does partial discharge mean equipment failure is imminent?
No. Discharge activity confirms that an insulation defect exists, but the time between first detection and failure varies widely with the defect type and operating conditions. Trending activity over successive tests, rather than a single reading, shows how quickly the condition is developing.
What is apparent charge in picocoulombs?
Apparent charge is the quantity IEC 60270 uses to characterize partial discharge, expressed in picocoulombs (pC). It represents the charge measurable at the equipment terminals that corresponds to the discharge inside the insulation. It indicates discharge intensity and allows comparison between tests, though it is not the true charge at the defect site itself.
Can partial discharge testing be performed while equipment is in service?
Yes. Online methods measure discharge on energized equipment: transient earth voltage (TEV) sensing on switchgear, high-frequency current transformer (HFCT) testing on cable earth connections, ultrasonic and acoustic detection, and UHF sensing. Conventional IEC 60270 measurement is typically an offline procedure that requires the asset to be de-energized.
How often should partial discharge testing be scheduled?
Practice varies with asset class, voltage level, and criticality. A common approach is a baseline survey of medium- and high-voltage assets, followed by periodic retests at a fixed interval. Assets that show rising discharge activity get shorter intervals, so the trend itself guides the schedule.
The Bottom Line
Partial discharge is an early, measurable warning that insulation in switchgear, transformers, machine windings, or cables is deteriorating. IEC 60270 testing quantifies the condition offline in picocoulombs, while TEV, HFCT, ultrasonic, and UHF methods keep energized assets under observation. Treat the results as a trend, because a defect whose activity grows between tests deserves action before a stable one does. Baseline the MV/HV fleet, retest on a cycle, and pair electrical PD testing with thermal and acoustic inspection for the fullest picture of insulation health.
Monitor Electrical Asset Health Continuously
Tractian's multimodal asset health platform combines mechanical, electrical, and operational signals in one system, giving maintenance teams a single view of critical equipment condition.
Explore Condition MonitoringRelated terms
Kaizen: Definition
Kaizen is the Japanese philosophy of continuous, incremental improvement involving every person in an organization. Learn the PDCA cycle, Kaizen events, the eight wastes in maintenance, and how Kaizen drives OEE improvement.
Failure Finding Interval (FFI): Definition
A failure finding interval (FFI) sets how often dormant protective equipment must be tested. Learn the formula, how FFI differs from PM intervals, and how it is applied in RCM programs.
Failure Analysis: Definition
Failure analysis is the systematic process of investigating why an asset failed. Learn the types (RCA, FMEA, FRACAS, FTA), the six-step process, and how findings improve maintenance strategy.
Gemba Walk: Definition
A Gemba walk is a lean practice where leaders observe work at its actual location to find improvement opportunities. Learn the steps, checklist, common mistakes, and how it applies to maintenance.
Functional Failure: Definition
A functional failure is the inability of an asset to meet its required performance standard. Learn how RCM uses functional failures to drive maintenance strategy and prevent operational losses.