Transformer Oil AC Dielectric Loss Testing (Dissipation Factor / Tan Delta)

Transformer oil AC dielectric loss testing measures the dielectric dissipation factor (tan δ, or power factor) of insulating oil to assess its dielectric integrity, contamination level, and aging condition. Conducted in accordance with internationally recognized standards such as IEC 60247 and ASTM D924, this test is a cornerstone of quality control for new oil and condition monitoring for in-service transformers. This article explains the principle, sample requirements, testing items, measurement methods, and typical applications of AC dielectric loss measurement for transformer oils.

Technical Overview

Transformer oil (mineral insulating oil) performs two critical functions in power equipment: electrical insulation and heat dissipation. When an alternating voltage is applied, an ideal dielectric stores energy capacitively and returns it to the circuit, dissipating no power. Real insulating liquids, however, always lose a small amount of energy as heat due to ionic conduction, dipole relaxation, and interfacial polarization. The AC dielectric loss of transformer oil quantifies this energy dissipation and is expressed as the dielectric dissipation factor, tan δ, defined as the ratio of the resistive (loss) current component to the capacitive (charging) current component at power frequency.

tan δ is a dimensionless quantity, commonly reported as a decimal (for example, 0.005) or as a percentage (0.5%). For the very small loss angles encountered in good-quality oil, the dissipation factor and the power factor are numerically almost identical, and the two terms are frequently used interchangeably. The measurement is highly sensitive to polar and conductive species in the oil, which makes it an early and reliable indicator of insulation degradation.

Elevated dielectric loss is typically caused by:

• Moisture dissolved or emulsified in the oil
• Oxidation products such as acids, peroxides, aldehydes, and sludge precursors
• Polar contaminants introduced during manufacturing, transport, or maintenance
• Suspended conductive particles, including metal fines and carbon
• Thermal and electrical aging by-products from solid cellulose insulation
• Mixing of incompatible oils or contamination with other fluids

Because dielectric loss increases sharply with temperature and with degradation, the standardized measurement is performed at an elevated reference temperature (typically 90 °C) at power frequency (50/60 Hz). The results are used to accept new oil, trend oil aging over the service life of a transformer, and verify the effectiveness of drying, degassing, or regeneration treatments.

Testing Samples

New (Unused) Mineral Insulating Oil

Samples of fresh oil taken at refineries, blending facilities, storage tanks, tanker deliveries, or equipment factories prior to filling. Verification against the delivery specification (for example, IEC 60296 or equivalent national specifications) confirms that the oil has not picked up moisture or polar contaminants during production, packaging, and transport.

In-Service Oil from Operating Equipment

Oil drawn from power transformers, instrument transformers, reactor coils, bushings, oil-filled cables, and on-load tap changers during routine maintenance or diagnostic investigations. Sampling follows standard practice (for example, IEC 60475) using clean, dry, sealed containers, and the sample is protected from light, air, and moisture until analysis. In-service results are compared with previous values to trend the aging condition of the oil-paper insulation system.

Regenerated, Reclaimed, or Treated Oil

Oil that has passed through vacuum dehydration, degassing, filtration, or full regeneration (for example, fuller's earth treatment) must be retested to confirm that polar oxidation products and moisture have been removed and that the dielectric loss has returned to an acceptable level before the oil is returned to service.

Alternative Insulating Liquids

Natural esters, synthetic esters, and silicone liquids used in fire-safe and environmentally preferred transformers can also be evaluated, using the appropriate product standards and grade-specific limit values, since the test principle is identical for all insulating liquids.

Testing Items

Core and related test items offered for transformer oil dielectric characterization include:

• Dielectric dissipation factor (tan δ) at power frequency, 50/60 Hz, measured at 90 °C in accordance with IEC 60247 / EN 60247 and equivalent national adoptions such as GB/T 5654
• Power factor (dissipation factor) at 25 °C or 100 °C in accordance with ASTM D924
• Relative permittivity (dielectric constant, εr) determined with the same cell and bridge configuration
• DC volume resistivity as a complementary conduction measurement frequently combined with tan δ testing
• Multi-temperature dielectric loss characterization for research or troubleshooting purposes

Testing Methods

Reference Standards

• IEC 60247 / EN 60247 – Insulating liquids: measurement of relative permittivity, dielectric dissipation factor (tan δ) and DC resistivity
• ASTM D924 – Standard test method for dissipation factor (or power factor) and specific resistivity of electrical insulating liquids
• GB/T 5654 – Chinese national method for relative permittivity, dielectric dissipation factor and DC resistivity of liquid insulating materials
• IEC 60296 – Specification for unused mineral insulating oils for transformers and switchgear (acceptance limits)
• IEC 60422 – Mineral insulating oils in electrical equipment: supervision and maintenance guidance (in-service evaluation)
• IEC 60475 – Method of sampling liquid dielectrics

Test Apparatus

The measurement employs a precision three-terminal (guarded) test cell fitted with polished gold-plated or stainless steel electrodes and a defined electrode gap, usually around 2 mm. The guard electrode eliminates surface leakage and stray capacitance so that only the current through the oil sample is measured. The cell is placed in a temperature-controlled oil bath capable of maintaining the reference temperature, typically 90 °C ± 0.5 °C or better, because tan δ is strongly temperature dependent. Dissipation is measured with an AC bridge (classical Schering bridge or transformer-ratio-arm bridge) or, in modern laboratories, with an automated digital dielectric analyzer that balances and reads tan δ directly at the selected voltage and power frequency. The applied stress is standardized, typically of the order of 0.5 to 1 kV/mm for IEC-based measurements.

Measurement Procedure

• Thoroughly clean and dry the test cell using suitable solvents and warm dry air; verify the cell constant and residual loss with an empty-cell check before each series.
• Pour the representative sample gently into the cell, avoiding air bubbles, dust, and moisture pickup; fill to the marked level and, where required, protect the surface with dry nitrogen.
• Place the cell in the bath and allow sufficient temperature stabilization time until the entire sample reaches the reference temperature.
• Apply the specified AC test voltage at 50/60 Hz, allow a brief voltage conditioning period, then record the tan δ reading; repeat and average multiple readings for reproducibility.
• Record the test temperature, frequency, applied voltage gradient, and cell identification; report tan δ to four decimal places together with the test conditions.
• Clean the cell immediately after testing to prevent cross-contamination ("memory effect") between samples.

Quality Control and Interference Management

Accredited third-party laboratories control the principal error sources of the method: moisture absorption during sampling and cell filling, particulate contamination, residual cleaning solvent, electrode fouling from previous samples, and temperature gradients within the cell. Participation in interlaboratory comparisons, verification of the bridge with reference capacitors, and replicate determinations provide confidence in the reported values. Because the measurement is extremely sensitive, careful technique is essential; a sample mishandled for even a short time in a humid environment can show a falsely elevated loss factor.

Applications

• Acceptance testing of new oil deliveries against purchase specifications before filling into transformers, reactors, bushings, or switchgear
• Factory quality control by transformer manufacturers during equipment production and commissioning
• Routine condition monitoring of in-service transformers as part of an integrated oil analysis program alongside dissolved gas analysis, moisture, acidity, and breakdown voltage testing
• Verification of maintenance treatments such as vacuum dehydration, degassing, filtration, or full oil regeneration
• Diagnosis of suspected contamination, oil mixing, or accelerated thermal-oxidative aging
• Failure investigations and forensic analysis following equipment incidents
• Research and development support for insulating fluid producers, additive evaluation, and studies of oil-paper insulation aging
• Input data for asset health indexing, remaining life assessment, and repair-versus-replace decisions on aging transformer fleets

As a general guideline, unused mineral transformer oil is expected to show a very low dissipation factor at 90 °C, with common specifications limiting tan δ to values on the order of a few thousandths or lower, while premium grades routinely achieve substantially lower figures. For in-service oil, condition-assessment guides such as IEC 60422 define classification bands (good, fair, poor) that become progressively stricter with increasing equipment voltage class; utilities normally establish their own trending limits based on voltage level, load criticality, and historical data. Any significant step increase between consecutive measurements warrants investigation even when the absolute value remains within limits.

FAQ

What is the difference between dissipation factor (tan δ) and power factor?

Both describe the same energy loss in the dielectric. The dissipation factor is the tangent of the loss angle, while the power factor is the cosine of the phase angle. For good insulating oils the loss angle is very small, so the two values are numerically nearly equal; IEC-based practice typically reports tan δ, whereas ASTM-based practice often reports power factor.

Why is the measurement performed at 90 °C?

Dissipation factor rises steeply with temperature, so measuring at a standardized elevated temperature maximizes sensitivity to polar contaminants and aging products, reflects the thermal environment inside operating equipment, and eliminates ambiguity caused by ambient temperature variation between laboratories and test dates.

What causes the dissipation factor of transformer oil to increase?

The most common causes are moisture ingress, oxidation and thermal aging products (acids, sludge precursors), polar contamination, conductive particles, and degradation of the solid cellulose insulation. A rising tan δ trend is therefore treated as an early warning of insulation system deterioration.

How much sample is required and how should it be taken?

A typical three-terminal cell requires only a few tens of milliliters of oil, although laboratories generally request a larger volume (for example 250 mL to 1 L) to allow combined testing. Sampling should follow IEC 60475 using clean, dry, sealed glass or metal containers, with the sample protected from moisture, light, and air, and delivered to the laboratory promptly.

How often should dielectric loss testing be carried out?

For critical high-voltage transformers, tan δ is commonly included in annual or biennial oil test programs, with shorter intervals when a degrading trend is observed. Testing is also performed at commissioning, after any oil handling or treatment, and as part of post-failure investigations.

Can a high dissipation factor be corrected?

In many cases yes. If the elevation is caused by moisture or gases, vacuum dehydration and degassing will restore the value. If it is caused by oxidation products, full regeneration using adsorbent treatment can remove the polar species. Retesting after treatment confirms whether the target value has been achieved.