Partial Discharge Testing of Transformer Busbar Terminals

Partial discharge (PD) testing of transformer busbar terminals is a critical insulation quality control and diagnostic procedure used to detect hidden dielectric defects such as voids, inclusions, and sharp-edge corona in cast-resin busbars and terminal connections. Performed in accordance with IEC 60270 and the IEC 60076-11 series, the test measures apparent charge in picocoulombs to verify that insulation systems meet strict acceptance limits before transformers enter service. This article explains the principles, test samples, measured parameters, methods, and applications of PD detection for transformer busbar terminals.

Technical Overview

Partial discharge (PD) is a localized dielectric breakdown that occurs within or on the surface of an insulation system without completely bridging the electrodes. In transformers, busbar terminals and their supporting insulation are among the most stress-concentrated regions of the assembly, because bolted joints, sharp conductor edges, insulation interfaces, and cast-resin transitions create localized electric field enhancement. Micro-voids, gas pockets, metallic particles, or surface contamination introduced during manufacturing can act as PD initiation sites.

Although individual discharge events are extremely small, sustained PD progressively erodes insulation through ionic bombardment, chemical degradation, and electrical treeing. Over months or years of operation, this leads to tracking, carbonized channels, and ultimately catastrophic dielectric failure at the terminal region. For this reason, PD measurement is recognized as the most sensitive quality and condition indicator for solid-insulated busbar systems.

PD detection on transformer busbar terminals is primarily governed by IEC 60270 (High-voltage test techniques – Partial discharge measurements), supplemented by IEC 60076-11 for dry-type transformers, IEC 60137 for bushings, and the corresponding GB/T 1094.11 national standard series. In these standards, discharge activity is quantified as apparent charge, expressed in picocoulombs (pC), and compared against defined acceptance limits, typically 10 pC or less at the specified measuring voltage for dry-type insulation systems.

Testing Samples

Cast-Resin Insulated Busbars and Terminal Boards

Epoxy or polyurethane cast-coil busbar sections, including straight segments, elbows, T-branches, and terminal boards removed from dry-type transformers, are tested as discrete components. These samples most frequently contain voids or interfacial defects at conductor edges where casting pressure and resin flow were non-uniform.

Bolted Terminal Connections with Insulating Covers

High-current bolted joints fitted with heat-shrink sleeves, epoxy caps, insulating boots, or mud-type filling compounds are evaluated to confirm that field grading at the joint geometry does not generate discharge above the specified limit.

Bushing Terminals and Cable Connection Assemblies

Medium-voltage bushings, plug-in cable connectors, and stress-cone terminations attached to busbar outlets are tested to verify interface quality between the bushing, connector, and busbar insulation.

Complete Transformer Assemblies

For routine and type testing, the full transformer is energized so that PD from windings, busbar supports, and terminal hardware is measured together. Samples must be clean, dry, and stabilized at ambient conditions, with unprotected live edges temporarily shielded to avoid external corona masking internal defects.

Testing Items

The following parameters are recorded during a typical PD test campaign on busbar terminals:

• Apparent charge magnitude (pC) at the specified measuring voltage
• Partial discharge inception voltage (PDIV), the lowest voltage at which sustained PD appears
• Partial discharge extinction voltage (PDEV), the voltage at which PD ceases when the voltage is reduced
• Background noise level of the test circuit, which must be well below the specified limit
• Discharge behavior during the complete voltage withstand sequence, including any trend of increasing activity
• Optional phase-resolved PD pattern analysis for defect classification, and localization results when UHF or acoustic sensors are deployed

Testing Methods

Conventional Electrical Method per IEC 60270

The reference method couples discharge pulses through a coupling capacitor and coupling device (quadrupole) into a calibrated PD detector. The circuit is calibrated by injecting a known charge across the test object terminals, establishing the scale factor in picocoulombs. Measurements may use a direct circuit, with the coupling capacitor in parallel with the test object, or an indirect circuit exploiting the transformer bushing tap. Conducted in a shielded high-voltage laboratory with carefully graded HV leads and corona-free terminations, this method provides quantitative, standards-compliant results.

Standard Voltage Sequence

For routine testing of dry-type transformers and their busbar systems, a typical sequence defined in IEC 60076-11 consists of raising the voltage to a pre-stress level (for example 1.8 Um held for approximately 30 seconds), then reducing it to the measuring voltage of 1.1 Um, which is maintained for about 3 minutes while the apparent charge is recorded. The measured PD level must remain at or below the acceptance criterion, commonly 10 pC, with no sustained upward trend.

Ultra-High-Frequency (UHF) Method

UHF sensors installed on dielectric windows, sensor ports, or near terminal boxes detect electromagnetic radiation from discharges in the 300 MHz to 1.5 GHz band. Because the method is largely immune to external corona noise, it is well suited to on-site commissioning tests and permanent online monitoring of in-service transformers.

Acoustic Emission Method

Piezoelectric sensors mounted on the tank or enclosure capture acoustic pulses generated by PD events, typically in the 20 kHz to 300 kHz range. Time-of-arrival triangulation with multiple sensors allows approximate localization of discharge sources at specific busbar terminals, which is valuable for targeted maintenance.

High-Frequency Current Transformer (HFCT) Method

A clamp-on HFCT placed around cable sheath connections or grounding leads captures high-frequency discharge currents, typically in the 3 MHz to 30 MHz band. It is a convenient non-invasive technique for field surveys and periodic condition checks on energized installations.

Test Procedure and Quality Controls

• Visual inspection and dielectric cleaning of terminal assemblies, with temporary corona shielding applied where necessary
• Circuit calibration and verification that background noise is typically below 10 percent of the acceptance limit
• Application of the programmed voltage sequence with continuous PD recording
• Evaluation against the specified pC limit, with investigation of PDIV and PDEV margins when required
• Issuance of a test report including circuit details, calibration data, voltage sequence, measured values, and environmental conditions

Applications

• Routine factory acceptance testing of dry-type and cast-resin transformers prior to shipment
• Type testing of new busbar profiles, insulating enclosures, and terminal hardware designs
• Incoming inspection of cast-resin busbar components at transformer assembly plants
• Commissioning verification after transport, installation, or busbar retrofit work
• Condition-based maintenance and online monitoring of operating units in substations, data centers, and industrial plants
• Failure analysis and root-cause investigation following terminal insulation breakdowns
• Research and development of epoxy formulations, casting processes, and field-grading structures

FAQ

Why are busbar terminals particularly prone to partial discharge?

Terminals combine bolted joints, conductor burrs, right-angle edges, and multiple insulation interfaces. These features locally intensify the electric field, so even small manufacturing imperfections such as voids, trapped gas, or surface contamination can initiate discharge at operating voltage.

What acceptance criterion is typically applied?

For dry-type transformer insulation systems tested per IEC 60076-11 or the equivalent GB/T 1094.11 standard, the apparent charge at the measuring voltage of 1.1 Um must generally not exceed 10 pC. Specific product standards or customer specifications may impose stricter limits.

How sensitive is the measurement?

With proper calibration, a low-noise shielded circuit, and correct test object grounding, conventional IEC 60270 systems can reliably resolve discharge levels of a few picocoulombs, which is sufficient to detect void-type defects well before service failure.

Can PD testing be performed on an energized, in-service transformer?

Yes. UHF, acoustic emission, and HFCT techniques are non-invasive and are widely used for online monitoring and periodic field surveys, although quantitative picocoulomb readings in accordance with IEC 60270 require a de-energized laboratory or on-site test circuit.

What causes false indications during testing?

External corona from unshielded HV leads, poor grounding loops, fluorescent lighting, and nearby switching operations can all inject signals into the measuring band. Good laboratory practice, phase-resolved pattern analysis, and sensor fusion between electrical and acoustic methods help distinguish genuine internal PD from interference.

Is a laboratory test report transferable between manufacturers and test houses?

Yes, provided the test is performed by accredited third-party laboratories or manufacturer laboratories using calibrated IEC 60270 instrumentation, documented voltage sequences, and traceable calibration, ensuring results are comparable and acceptable to utilities and certification bodies worldwide.