Home › Transformer Testing Guide

Power Transformer Testing & Commissioning: The Complete Field Guide

A power transformer is usually the single most expensive asset in a substation, and the one with the longest replacement lead time. A transformer that fails on energisation — or worse, a few weeks into service — costs far more than the testing programme that would have caught the problem. That is why transformer testing and commissioning is a discipline of its own: a structured sequence of inspections and electrical tests that prove the machine arrived, was installed, and can be energised safely and reliably.

This guide explains the tests the way they are performed on site: what each test measures, why it matters, how it is done in general terms, and what a healthy result looks like. It follows the structure used in international practice under standards such as the IEC 60076 series and IEEE C57 guides. Always work to your project specification, the manufacturer's instruction manual and your local safety rules — those documents override any general guide, including this one. For the numbers side of the job, keep our transformer calculators open in another tab.

Safety first: before any test begins

Every transformer test programme starts with safety, not with a test set. The transformer must be isolated from all sources of supply, earthed, and issued under a permit-to-work system. Windings can store charge after insulation testing, so discharge and earthing after each high-voltage test is a habit, not a suggestion. Oil work brings its own hazards: hot oil, spill control, fire watch during processing, and strict no-smoking discipline around the tank. Lifting and moving radiators, bushings and conservators requires certified lifting plans. A test result is never worth an injury, and experienced commissioning engineers treat the safety briefing as part of the test itself.

The three families of transformer tests

Routine tests

Routine tests are performed on every transformer at the factory before dispatch: winding resistance, voltage ratio and phase displacement, short-circuit impedance and load loss, no-load loss and current, insulation tests, and induced-voltage withstand. On site, the commissioning team repeats the practical subset of these tests to prove that transport and installation did not change anything. Comparing site results with the factory test certificate is the single most powerful diagnostic habit in commissioning: a healthy transformer should reproduce its factory fingerprint.

Type tests

Type tests are performed once on a representative unit of a design — temperature-rise testing and lightning-impulse testing are the classic examples. Site engineers rarely repeat type tests, but they read the type-test certificates during documentation review to confirm the design has been proven for the duty the transformer will see.

Site (special and diagnostic) tests

Site testing also includes diagnostic tests used when condition matters as much as correctness: insulation power factor (tan delta), capacitance measurement, frequency response analysis to detect winding movement after transport, and dissolved gas analysis of the oil. These give a baseline — a fingerprint of the transformer in its installed condition — against which every future maintenance test will be compared.

Step 1: Receipt inspection and mechanical checks

Testing begins the moment the transformer arrives. Inspect for transport damage: check the impact recorder if one is fitted, look for dented radiators, cracked bushings, oil leaks at every gasket and valve, and confirm the nitrogen or oil pressure state matches the transport record. Verify the nameplate against the approved drawings — rating, voltages, vector group, impedance, cooling class — because a wrong nameplate discovered after erection is a programme disaster. Check that all accessories are present: conservator, silica gel breather, Buchholz relay, temperature gauges, pressure relief device, valves, and the tap changer with its drive mechanism.

Step 2: Insulation resistance and polarisation index

The insulation resistance (IR) test is usually the first electrical test, performed with a megohmmeter — see our guide to choosing an insulation tester for instrument selection. Resistance is measured between each winding and earth, and between windings, typically with a 5 kV test set on high-voltage equipment. Because insulation resistance is strongly affected by temperature and surface moisture, results are always recorded with winding and ambient temperature and compared like-for-like.

The polarisation index (PI) removes some of that guesswork: it is the ratio of the 10-minute resistance reading to the 1-minute reading. Dry, healthy insulation keeps absorbing charge, so its apparent resistance rises steadily and the PI is high — a value around 2 or above is generally considered good for oil-impregnated insulation, while a PI close to 1 suggests moist or contaminated insulation that needs drying or oil treatment before any high-voltage test is attempted. Never energise a transformer whose insulation condition you doubt: the IR and PI results are the gate through which everything else must pass.

Step 3: Turns ratio and vector group verification

The turns ratio test proves that the ratio of primary to secondary voltage matches the nameplate on every tap position. A ratio bridge or automatic transformer turns ratio (TTR) set applies a known voltage to one winding and measures the induced voltage in the other, giving ratio and ratio error directly. Results are usually expected well within half a percent of the calculated ratio; a larger error points to shorted turns, a wrong tap connection, or an internal connection error that must be found before energisation. Use our turns ratio calculator to pre-calculate the expected secondary voltage for each tap so the site readings can be checked on the spot.

Vector group verification confirms the phase displacement between the high-voltage and low-voltage windings — for example, that a unit nameplated as a delta-star group really is connected that way internally. Getting this wrong and paralleling or energising against a different phase relationship can be catastrophic, so the check is done carefully with the standard clock-method connections, together with polarity and phase-sequence confirmation.

Step 4: Winding resistance measurement

Winding resistance is measured with a DC test set — a low-resistance ohmmeter or dedicated winding resistance tester — injecting current through each winding and measuring the voltage drop. The test detects broken strands, poor brazed joints, loose connections and tap changer contact problems. Readings between phases should be closely balanced (within a few percent), and the values are corrected to a reference temperature for comparison with factory results. One practical warning every field engineer learns: a transformer core stores serious magnetic energy during this test, and disconnecting test leads while current is flowing can produce a dangerous voltage spike. Follow the test set's demagnetisation and discharge sequence, every time.

Step 5: Magnetising current and no-load behaviour

Measuring the magnetising (excitation) current at a reduced applied voltage gives a sensitive indication of core condition and shorted turns. The three phase readings follow a recognisable pattern in a healthy three-limb core, and a phase that suddenly disagrees with its sisters — or with the factory fingerprint — deserves investigation. This test is quick, cheap and surprisingly revealing, which is why experienced commissioning engineers never skip it.

Step 6: Transformer oil testing

In an oil-filled transformer the oil is both coolant and insulation, so its condition is tested as rigorously as the windings. The key site tests are:

• Breakdown voltage (BDV): a sample is drawn from the bottom drain valve and sparked across a standard electrode gap in a test cell. Repeated breakdowns are averaged; clean, dry oil withstands a high voltage, while water and particles collapse the value. Low BDV after transport or storage usually means the oil needs filtration and vacuum processing.
• Moisture content: measured in parts per million, typically by Karl Fischer titration in a laboratory or a portable instrument. Water in oil accelerates paper insulation ageing dramatically, so moisture limits tighten as voltage class rises.
• Dissolved gas analysis (DGA): a laboratory technique that measures gases dissolved in the oil. Hydrogen, acetylene, ethylene and carbon oxides each tell a story — overheating, partial discharge or arcing — making DGA the most powerful early-warning tool in transformer condition monitoring.
• Acidity, interfacial tension and colour: slower condition indicators, mostly used on in-service units to judge oil ageing and sludge formation.

Sampling technique matters: a badly drawn sample gives a meaningless result. Bottles must be clean, dry and filled with minimal air contact, and sampling valves flushed before the sample is taken.

Step 7: Insulation power factor (tan delta) and capacitance

Where specified, tan delta testing measures the dielectric losses of the winding insulation and bushings. A rising loss angle compared with factory or previous results indicates moisture or deterioration in the insulation system. Bushing tan delta and capacitance are particularly valuable because condenser bushings can deteriorate gradually — and this test catches them before a bushing failure takes the whole transformer with it.

Step 8: Protection, alarms and auxiliary systems

A transformer is protected by its relays and its own devices, and commissioning must prove every one of them works: Buchholz relay alarm and trip stages, oil and winding temperature alarm and trip contacts, pressure relief operation, low oil level, cooling fans and pumps starting on temperature and on load, and the tap changer's full range of operation — local, remote and automatic control where fitted. Secondary injection and functional trip testing prove the chain from device contact to circuit breaker open. Marshalling wiring, cable insulation and terminal tightness are checked as part of this stage, and every alarm is proven at the control panel or substation automation system, not just at the device.

Step 9: Earthing, cabling and pre-energisation checks

Before energisation the checklist closes out the physical installation: tank earth connections made with the specified conductor and measured earth resistance, neutral earthing as per design, cable terminations torqued and tested, CT circuits proven with correct polarity and ratio, surge arresters connected, clearances verified, oil level correct at the ambient temperature, silica gel active, all valves in their correct service positions, and any transport locking or shipping devices removed. A surprising number of energisation failures are not electrical test failures at all — they are a closed valve, a shipping pin left in a relay, or a fan wired for the wrong rotation. The checklist exists because memory alone is not a system.

Step 10: Energisation and on-load checks

Energisation is done from one side only, with protection in service and personnel clear. Expect a magnetising inrush — a brief, heavy current as the core magnetises — and listen to the transformer: a healthy unit settles into a steady hum, while unusual crackling or violent noise means switch off and investigate. Immediately after energisation, verify voltages and phase sequence on the secondary side, then apply load gradually. On-load checks include current balance, temperature behaviour, cooling operation under real load, tap changer operation under load where applicable, and — after a period of service — a follow-up oil sample for DGA to establish the in-service baseline.

Common problems found during commissioning

Finding	Likely cause	Typical action
Low insulation resistance / PI	Moisture absorbed during transport or storage	Oil filtration and vacuum processing, or drying procedure per manufacturer
Ratio error on one tap only	Tap changer contact or lead connection problem at that tap	Inspect tap changer connections and contact condition
One phase winding resistance high	Poor joint or damaged connection in that phase	Re-check connections; escalate to manufacturer if internal
Low oil BDV	Wet or contaminated oil	Filter and vacuum-process oil, re-test until values recover
Alarm not reaching control room	Marshalling wiring or auxiliary relay fault	Point-to-point wiring check and loop test

If you are building your instrument kit for this work, our electrical test tools buying guide explains how to choose insulation testers, multimeters and thermal cameras for transformer and substation work. And once the transformer is in service, your documentation is what proves all of this happened — our AI tools guide shows how engineers are speeding up test report writing without cutting corners.

Frequently asked questions

Written by Abdul Wadood, electrical site engineer — transformer execution, testing and commissioning. Educational content only; always follow your project specification and manufacturer instructions.