On-Site Dielectric Loss Factor (Tan Delta) Testing: Principles, Methods, and Applications

On-site dielectric loss factor (tan delta) testing is one of the most effective diagnostic techniques for evaluating the insulation condition of high-voltage electrical equipment such as transformers, bushings, instrument transformers, and capacitive apparatus. By measuring the dissipation factor and capacitance of insulation systems directly in the field, maintenance teams can detect moisture, aging, and contamination before catastrophic failures occur. This technical guide explains the measurement principles, tested equipment types, field test procedures, interference suppression techniques, and result assessment criteria for dielectric loss factor field testing.

Technical Overview

The dielectric loss factor, expressed as tan delta (tan δ), is the tangent of the dielectric loss angle, defined as the ratio of the resistive (loss) component of leakage current to the capacitive component of current flowing through an insulation system under AC voltage stress. Because dielectric loss is proportional to the active power dissipated as heat within the insulation, tan delta serves as a sensitive indicator of the overall insulation quality of high-voltage apparatus.

Ideal insulation behaves as a nearly perfect capacitor in which current leads voltage by 90 degrees. Real insulation contains polarization losses and conduction losses that shift the current phase slightly below 90 degrees; this angular deviation is the loss angle delta. The dielectric loss power can be expressed as P = U² × ω × C × tan δ, where U is the applied voltage, ω is the angular frequency, and C is the equivalent capacitance. An increase in tan delta therefore directly reflects an increase in energy dissipation within the insulation structure.

Field tan delta testing is classified as a non-destructive diagnostic method. Unlike withstand voltage tests, which mainly reveal concentrated defects such as gaps and punctures, dielectric loss measurement is highly sensitive to distributed, generalized insulation deterioration caused by moisture ingress, oil aging, insulation contamination, and long-term thermal stress. The test is typically performed at power frequency (50 Hz or 60 Hz) with a test voltage of approximately 10 kV, although the voltage may be adjusted for lower-rated equipment.

• Detects distributed insulation defects such as overall moisture, aging, and pollution that conventional insulation resistance or withstand tests may miss
• Simultaneously provides capacitance values, enabling detection of shorted capacitor elements, winding deformation, and structural displacement
• Serves as a key commissioning and preventive maintenance criterion in international and national standards
• Supports trending analysis when compared against factory data and historical test records

Testing Samples

On-site dielectric loss testing applies to virtually all major capacitance-graded and oil-paper insulated equipment in a substation or power plant. The most commonly tested apparatus are described below.

Power Transformers and Shunt Reactors

For oil-immersed power transformers, the tan delta and capacitance of each winding arrangement is measured, typically between the high-voltage winding and ground, between the low-voltage winding and ground, and between the high-voltage and low-voltage windings. Because the measurement represents a composite of oil, solid insulation, and bushings, results are compared against factory values and previous field records. For equipment rated 66 kV and above, typical acceptance values at 20 °C are on the order of 0.6% to 0.8%, with higher limits for lower voltage classes. A significant increase between consecutive tests indicates progressive insulation deterioration, moisture absorption, or oil degradation.

High-Voltage Bushings

Condenser-type bushings, including oil-impregnated paper (OIP), resin-impregnated paper (RIP), and resin-impregnated synthetic bushings, are among the most critical test objects. The test is performed on the main insulation (C1), measured between the center conductor and the test or measuring tap, and on the tap insulation (C2), measured between the tap and the grounded flange. New OIP bushings generally exhibit tan delta below 0.7% at 20 °C, and values exceeding approximately 1.0% to 1.5% in service warrant further investigation. Rising tan delta in bushings is a classic early warning of moisture ingress, partial discharge damage, or deterioration of the condenser core.

Instrument Transformers

Current transformers, voltage transformers, and transformer winding assemblies are tested to assess the condition of their oil-paper or cast resin insulation systems. Measurement of tan delta and capacitance between primary and secondary windings, and between each winding and ground, reveals moisture absorption and insulation aging. These devices often operate for decades, making periodic dielectric loss measurement an essential aging monitor.

Capacitive Apparatus

Coupling capacitors, capacitor voltage transformers (CVTs), capacitor divider units, breaker grading capacitors, and capacitor banks are routinely tested. Healthy oil-paper or film-paper dielectric units typically show tan delta values below 0.2% to 0.5% depending on design and voltage class. An abrupt increase in tan delta accompanied by a decrease in capacitance indicates failed shorted elements, while rising tan delta at stable capacitance indicates dielectric deterioration.

Power Cables and Rotating Machine Windings

For medium- and high-voltage XLPE and oil-paper cables, tan delta is often measured at very low frequency (VLF, typically 0.1 Hz) using dedicated test sets; healthy XLPE insulation exhibits tan delta values below approximately 0.4%, while water treed or degraded insulation shows significantly elevated values and strong voltage dependence. For generator and motor stator windings, the dissipation factor tip-up test, defined as the difference between tan delta measured at high and low test voltages, is performed according to IEC 60034-27 to evaluate the condition of groundwall insulation and the global void content.

Testing Items

A complete on-site dielectric loss test campaign normally includes the following measured and derived quantities:

• Tan delta (dissipation factor) of each insulation section at a standard test voltage, corrected to 20 °C
• Capacitance of each insulation section, compared against nameplate and previous values
• Tan delta versus voltage characteristics (tip-up curve) where required, to reveal voltage-dependent defects such as internal voids and partial discharge activity
• Trend analysis of tan delta and capacitance over successive maintenance cycles
• Comparison of field results with factory acceptance test data after correction for temperature and test configuration differences

Testing Methods

Measurement Principles

Classical dielectric loss measurement is based on the Schering bridge, in which the test object capacitance and the reference standard capacitor form two arms, and adjustable resistive and capacitive elements form the balancing arms. When the bridge is balanced, tan delta and capacitance are calculated directly from the balance settings. Modern field instruments use digital phase-sensitive sampling: the instrument measures the magnitude and phase of the current through the test object relative to the voltage across a low-loss reference capacitor, and computes tan delta and capacitance automatically with high resolution and repeatability.

Wiring Configurations

Three basic connection modes are used depending on the grounding arrangement of the specimen. In the direct or ungrounded specimen test (UST), one terminal of the test object is energized while the other terminal is connected to the measuring circuit; this mode is suitable for bushing C1 tests and ungrounded two-terminal apparatus. In the reversed or grounded specimen test (GST), the test object is energized while its grounded terminal returns the loss current to the bridge; this mode dominates field work because most installed equipment has one terminal permanently grounded. Guard techniques route unwanted leakage currents, such as surface leakage across a bushing, away from the measuring circuit so that only the insulation volume under evaluation contributes to the reading.

Anti-Interference Techniques

Field tests are frequently performed adjacent to energized switchgear and lines, producing electrostatic and electromagnetic interference at power frequency that can severely distort readings. The most effective modern countermeasure is frequency conversion: the instrument measures tan delta at two offset frequencies, for example 45 Hz and 55 Hz, and interpolates the result at 50 Hz, thereby rejecting interference concentrated at system frequency. Additional measures include heavy grounding of the test set, shielding of measurement leads, use of the bridge inversion or polarity reversal method to cancel residual errors, and the relative comparison method, in which identical apparatus on the same bus are measured under the same interference environment and their results are compared to each other.

Test Procedure and Temperature Correction

The specimen is first de-energized, isolated from all sources, and solidly grounded to discharge stored capacitive energy. The instrument is grounded, the high-voltage lead is supported clear of ground with an insulated pole, and the appropriate wiring mode is selected. After application of the test voltage, readings are taken, repeated, and averaged. Because tan delta of oil-paper insulation rises significantly with temperature, all results are converted to the reference temperature of 20 °C using standardized correction factors before comparison. Surface conditions also matter: bushing surfaces should be clean and dry, and shielding electrodes may be applied to eliminate surface leakage during humid weather. Immediately after the test, the specimen is again discharged and grounded before any reconnection.

Applications

On-site dielectric loss factor testing is applied across the entire life cycle of high-voltage equipment:

• Commissioning and handover acceptance testing of newly installed transformers, bushings, and capacitive apparatus before energization
• Periodic preventive maintenance testing performed annually or on multi-year cycles as required by maintenance regulations
• Condition-based assessment of aging equipment to support repair, refurbishment, or replacement decisions
• Post-fault and post-repair verification to confirm that moisture has been removed by drying or that insulation condition has been restored
• Pre-outage diagnostics to prioritize maintenance resources and reduce unplanned outages
• Baseline data capture for trending, enabling early detection of gradual deterioration long before any failure risk exists

FAQ

What is an acceptable dielectric loss factor value?

Acceptance limits depend on the equipment type, voltage class, and insulation technology. As general guidance at 20 °C, healthy oil-paper transformer insulation measures below approximately 0.8%, condenser bushing main insulation below 0.7% to 1.0%, and coupling capacitors below 0.2% to 0.5%. Equally important is the trend: a rapid increase relative to previous results is often a stronger warning than any single absolute number.

Why must field results be corrected to 20 °C?

Tan delta of oil-paper and resin insulation increases markedly with temperature, and the rate of increase itself grows as insulation degrades. Comparing measurements taken at different temperatures without correction would produce misleading conclusions, so all results are converted to the standard reference temperature of 20 °C using recognized correction tables.

Can testing be performed near energized equipment?

Yes. Modern frequency-conversion instruments measure at frequencies offset from the power system frequency and interpolate to the rated frequency, effectively rejecting interference from adjacent live apparatus. Careful grounding, lead routing, and shielding further improve accuracy in high-interference substations.

What test voltage and frequency are normally used?

For most apparatus, a power-frequency test voltage of 10 kV is standard, with reduced voltages applied to lower-rated insulation. For cable systems, very low frequency (0.1 Hz) tan delta testing is widely used because it permits testing with compact equipment. Tip-up tests on machine windings apply a stepped voltage sequence up to the rated voltage range.

How does moisture affect the measurement?

Water in insulation sharply increases both conduction and polarization losses, causing tan delta to rise substantially and become more temperature dependent. A tan delta value that doubles or triples compared with historical data is a strong indicator of moisture ingress and usually triggers drying, oil processing, or further laboratory analysis of the insulation.

How often should on-site dielectric loss testing be performed?

Most maintenance codes require annual measurement for critical equipment such as capacitive apparatus and instrument transformers, with transformer tests performed on one- to three-year cycles depending on voltage class, service age, and condition assessment history. Testing is also mandatory after major repairs and before initial energization of new installations.