Partial discharge (PD) testing in a factory acceptance test

Partial discharge testing is one of the type-specific acceptance tests that sit inside the broader factory acceptance test. For where the FAT fits in the procurement sequence, see the factory acceptance testing guide; for why a test passed in the factory is not the same as one passed on site, see FAT vs SAT. This page covers the PD test specifically: what it reveals, which equipment needs it, how it differs at FAT and SAT, and what a result a buyer can rely on actually records.

What a PD test can reveal that a withstand test does not

Routine withstand tests — the dielectric or “hipot” tests — answer one question: did the insulation survive a defined over-voltage for a defined time. They are pass/fail. They do not reveal incipient weakness: a void, a contaminant, a sharp edge or a delamination that is not yet failing but can degrade in service. Partial discharge is that earlier signal — a discharge that bridges only part of the insulation, measured as apparent charge in picocoulombs (pC) when the conventional IEC 60270 method is used. A unit can pass a withstand test and still show a PD level that indicates elevated insulation risk or a defect that warrants investigation. For high-voltage equipment, where an insulation failure is severe and not visible from the outside until it happens, PD measurement adds evidence of insulation quality that a withstand test alone does not provide.

Which equipment needs a PD test at FAT

PD measurement is not required on every order. It belongs where the equipment is high voltage, an insulation defect is a dominant failure mode, and the product standard or purchaser specification calls for it. Low-voltage equipment, mechanical packages and finished goods generally do not need it. The measurement method and acceptance basis differ by equipment type:

PD at FAT versus PD at SAT

At FAT, PD is measured in the factory or a qualified test laboratory, preferably in a controlled or shielded HV environment with calibrated instruments, a recorded background noise level and a defined voltage profile. This is often the cleanest and most repeatable measurement available before shipment, and the lowest-risk point to reject a defect — while the unit is still on the supplier’s floor and the relevant payment milestone is unmet.

At site, during SAT or commissioning, PD is often re-measured because transport, handling and installation can introduce new defects — though the site noise floor is higher and the measurement is harder. Where PD is specified, a factory PD pass is necessary but not sufficient for final site acceptance: it is the baseline the site result is read against. This is the same gap that separates any factory test from its site counterpart — see FAT vs SAT for why a passed factory test is not a passed site test.

Who provides the PD test equipment?

The PD measuring system is normally provided by the party running the test — the manufacturer’s HV laboratory, or a suitably qualified third-party test laboratory where the manufacturer cannot test to the required level. Sinospect does not sell or supply PD test instruments such as PD detectors or coupling capacitors. Its role is to specify the PD test in the FAT protocol, arrange it, witness the measurement and control the acceptance record — and, where Sinospect supplies the high-voltage equipment as principal, to treat the PD result as part of the QC gate before the relevant payment milestone is released.

What a credible PD result looks like

A PD result a buyer can rely on records, at minimum:

• the approved ITP or FAT protocol showing PD as a witnessed hold point;
• the standard and method used (e.g. IEC 60270) and the calibration of the PD measuring system, with traceable calibration certificates;
• the test circuit and the applied voltage profile and duration;
• the background or noise level in pC, measured with the test circuit energized so that the test equipment itself is verified not to be generating discharge — a result reported without its noise floor cannot be read;
• the measured apparent charge in pC against the limit set by the applicable product standard, purchaser specification or agreed protocol at the specified voltage;
• the discharge inception and extinction voltages (PDIV / PDEV) where the standard or protocol requires them;
• the serial number and nameplate of the actual unit tested, the test conditions, the laboratory identity and accreditation scope where a third-party lab is used, and the witness signatures with non-conformity disposition if anything failed.

Two red flags recur on China-sourced HV equipment: a “PD: OK” or “pass” line carrying no measured value, noise floor, calibration or test voltage — unless the applicable standard expressly permits a pass/fail statement — and a type-test report substituted where the contract or ITP requires PD evidence on the actual serial-numbered unit.

Frequently asked questions