Transformer Oil Testing: Dielectric Breakdown, IFT, Acid Number, and Moisture

Transformer insulating oil serves two functions: it provides electrical insulation between energized parts, and it transfers heat from the core and windings to the tank walls and radiators. As the oil ages and the transformer operates, both functions degrade gradually. Oxidation products accumulate, moisture enters from the atmosphere or from paper insulation degradation, and the dielectric strength diminishes. Periodic oil testing tracks this progression and identifies when the oil requires reconditioning, when the paper insulation has deteriorated, and when an active fault may be generating gases in the oil.

A standard transformer oil test routine includes dissolved gas analysis (DGA), dielectric breakdown voltage, interfacial tension (IFT), acid or neutralization number, moisture content by Karl Fischer titration, color, and specific gravity. Each parameter is independent and addresses a different aspect of oil and insulation condition. DGA is covered in a separate article on transformer dissolved gas analysis; this article covers the physical and chemical parameters that complete the oil condition picture.

Dielectric breakdown voltage (ASTM D877 and D1816)

Dielectric breakdown voltage is the most direct measure of the oil's electrical insulating capability. The test applies increasing AC voltage across a pair of electrodes immersed in the oil sample until the oil breaks down — an arc jumps the gap between the electrodes — and records the voltage at which breakdown occurs. The test is performed multiple times on the same sample (typically five or six breakdowns) and the results are averaged. Higher breakdown voltage means better insulating capability.

Two ASTM methods are used. ASTM D877 uses flat disk electrodes 25 mm in diameter with a 2.5 mm gap, energized at 3 kV per second until breakdown. ASTM D1816 uses VDE bowl-shaped electrodes with a smaller gap (1 mm or 2 mm) that produce a more uniform electric field and are more sensitive to contamination — particularly water droplets and particles — than the flat disk electrodes. D1816 gives lower absolute breakdown values than D877 on the same sample, so the two methods are not directly comparable; each has its own acceptance threshold.

IEEE C57.106 acceptance criteria for new transformer oil specify a minimum D877 breakdown voltage of 30 kV. For oil in service, the minimum acceptable value is typically 26 kV by D877, with values below 20 kV indicating the oil should be processed or replaced before the transformer returns to service. For critical large power transformers, many utilities apply more conservative limits — 35 kV minimum for new oil and 30 kV for in-service oil. A sudden drop in breakdown voltage from a prior test, even if the value is still above the minimum threshold, warrants investigation — it may indicate moisture ingress, contamination, or the beginning of a fault process generating conductive particles.

The primary cause of low breakdown voltage in transformer oil is moisture. Water reduces breakdown voltage dramatically — even a few hundred ppm of dissolved moisture can reduce breakdown voltage by 30–50% in mineral oil. A breakdown voltage test that falls below the minimum should always be paired with a Karl Fischer moisture measurement to determine whether moisture is the cause, because the remediation differs: moisture is removed by vacuum degassing and hot oil circulation, while particulate contamination requires filtration.

Interfacial tension (ASTM D971)

Interfacial tension measures the molecular force at the interface between transformer oil and distilled water, expressed in millinewtons per meter (mN/m) or dynes per centimeter. Fresh, uncontaminated mineral oil has a high IFT — typically 40–45 mN/m — because the non-polar oil molecules have little affinity for the polar water interface. As the oil oxidizes and polar oxidation products (acids, sludge precursors, and other polar compounds) accumulate, these polar molecules preferentially migrate to the oil-water interface and reduce the interfacial tension.

IFT is therefore an early and sensitive indicator of oil oxidation, often detecting oxidation products before they appear in the acid number test. A new transformer oil with IFT above 40 mN/m is in good condition. Values between 25–40 mN/m indicate early oxidation. Values below 25 mN/m indicate significant oxidation and suggest the oil should be reconditioned or replaced. Values below 18 mN/m indicate the oil is severely oxidized and sludge deposition on the core and windings may have already begun. IFT below 18 mN/m is a strong indicator that power factor testing and internal inspection should follow promptly, because sludge deposited on winding surfaces increases thermal resistance and drives accelerated aging of the paper insulation.

IFT and acid number together tell the oxidation story with more completeness than either alone. An oil with falling IFT but a still-low acid number is in the early oxidation stage where the oxidation products are primarily neutral polar compounds rather than acids — the oil is degrading but has not yet produced the acid species that accelerate paper insulation aging. An oil with both low IFT and high acid number has progressed further and the paper insulation is at greater risk.

Acid number / neutralization number (ASTM D974)

The acid number (also called the neutralization number or total acid number, TAN) measures the concentration of acidic compounds in the oil, expressed as the milligrams of potassium hydroxide (KOH) required to neutralize the acid in one gram of oil (mg KOH/g). Fresh mineral oil has a very low acid number, typically below 0.01 mg KOH/g. As the oil oxidizes, organic acids form as degradation products and the acid number rises.

Acid is damaging to transformer insulation because it attacks the cellulose paper insulation, hydrolyzing the polymer chains and reducing the degree of polymerization (DP) of the paper. Paper with reduced DP is mechanically weaker — it is more susceptible to damage from through-fault mechanical forces and vibration. This degradation is irreversible; unlike the oil, which can be reconditioned or replaced, paper that has been acid-degraded has permanently lost mechanical strength.

IEEE C57.106 acceptance criteria for new transformer oil require an acid number below 0.03 mg KOH/g. For in-service oil, the limit for continued service is typically 0.10 mg KOH/g; oil exceeding 0.20 mg KOH/g should be reconditioned; oil exceeding 0.40 mg KOH/g has severely degraded the insulation system and indicates the transformer may need an internal inspection as well as complete oil replacement. The rate of change in acid number between samples is as important as the absolute value — an acid number that has doubled in one year indicates accelerating oxidation and should trigger action even if the current value is still within limits.

Moisture content (Karl Fischer titration, ASTM D1533)

Water in transformer oil accelerates insulation aging, reduces dielectric breakdown voltage, and increases the risk of partial discharge in the insulation system. Karl Fischer titration is the standard method for measuring dissolved moisture in transformer oil, expressed in parts per million (ppm) by weight. The Karl Fischer reagent reacts stoichiometrically with water, and the endpoint is detected electrochemically with high precision.

The moisture in transformer oil is not independent of the moisture in the paper insulation — water distributes between the oil and paper according to an equilibrium relationship that depends on temperature. At high temperatures (the transformer at full load), the equilibrium shifts toward the oil, pulling moisture out of the paper into the oil. At low temperatures (the transformer off-load or lightly loaded), the equilibrium shifts back and moisture migrates from the oil into the paper. A moisture measurement on a sample taken when the transformer is cold and lightly loaded will therefore read higher than a sample taken from the same transformer at rated load. Correct interpretation requires knowing the oil temperature at the time of sampling.

IEEE C57.106 limits for moisture in oil vary by transformer voltage class. For transformers rated 69 kV and below, the limit for continued service is 35 ppm at operating temperature; for transformers rated 115–230 kV, the limit is 25 ppm; for 345 kV and above, 20 ppm. Exceeding these limits requires vacuum dehydration and hot oil circulation to remove moisture from both the oil and the paper. The relationship between oil ppm and paper moisture content — and the more diagnostic measure of relative saturation — is covered in detail in our article on transformer dew point and moisture testing.

Color (ASTM D1500)

Transformer oil color is assessed visually against a color scale from 0.5 (water-white) to 8.0 (dark brown). Fresh new oil is typically 0.5–1.0. As the oil ages and oxidizes, it darkens progressively through yellow, amber, and brown. Color is a crude but rapid indicator of oil condition — a transformer with dark brown oil (color 5 or above) has severely oxidized oil regardless of what the other parameters say, and the other parameters on the report will confirm the oxidation picture.

Color is not diagnostic by itself for specific failure modes, but a sudden darkening between sample intervals — particularly if accompanied by DGA anomalies — can indicate an internal fault event that has caused localized oil carbonization. Normal oxidative darkening is gradual; a step change in color warrants investigation.

Sampling procedure and its effect on results

All oil test results depend on a properly collected sample. The sample must be taken from the bottom drain valve after flushing several liters of oil through the valve to clear the valve body and connecting pipe of accumulated sediment and water. The sample container must be dry and clean; glass is preferred over plastic for most tests because plastic can leach compounds that interfere with IFT and acid number. The sample must not be exposed to air beyond what is necessary to fill the container, because atmospheric moisture and oxygen contaminate the sample and produce falsely elevated moisture and oxidation readings.

Southern Switch collects transformer oil samples per IEEE C57.106 sampling procedures and sends them to an independent laboratory (SDMyers or equivalent) for analysis. The laboratory report returns all parameters discussed here, plus DGA, specific gravity, flash point, and power factor on the oil. The complete report, interpreted alongside the electrical test results from power factor, insulation resistance, and DGA trending, gives a full picture of transformer insulation system condition that no single test can provide alone.