Insulation and dielectric
Cable tan delta testing measures the dielectric loss of a power cable's insulation while a very low frequency voltage is raised in steps. The level, the stability and the rise of the loss with voltage give a picture of overall insulation aging, so cables can be ranked and replaced by condition instead of by age.
Cable tan delta testing, also called dissipation factor or loss angle testing, is a diagnostic that measures how much energy the insulation of a power cable dissipates when an alternating voltage is applied. The tangent of the loss angle, written tan delta, is the ratio of the in-phase loss current to the capacitive current, and it is a dimensionless indicator of insulation quality. The test is applied mostly to medium-voltage extruded cables, such as those insulated with cross-linked polyethylene or ethylene propylene rubber, and to older paper-insulated cables. It is usually performed with a very low frequency sinusoidal source, because such a source is small enough to carry to the field yet can charge a long cable. The voltage is raised in steps related to the cable's rated phase-to-ground voltage, and tan delta is measured at each step. Unlike a pass-fail withstand test, the result is a set of numbers describing the overall condition of the whole cable segment, which makes the test useful for asset management.
Aging of cable insulation is usually gradual and widespread. Water trees, thermal degradation, contamination and poorly made accessories raise the dielectric loss of the insulation over years, and the cable eventually fails in service, often at a termination or splice or at a weak spot in the run. Unplanned cable failures cause outages, are expensive to repair and in some installations create safety risks. A test that only proves the cable survives a stress cannot rank cables by condition, and it can even damage insulation that is already weak. Tan delta testing is non-destructive when run within published voltage limits, and it identifies segments whose losses are rising before a failure occurs. This lets owners decide which circuits to replace, which to rejuvenate and which to leave alone. It is run on new cable and its accessories to confirm that installation caused no harm, and on a recurring schedule on aging circuits so that a trend can be established and the replacement budget can be directed to the worst cables first.
A perfect insulation behaves as a pure capacitor, drawing a current that leads the applied voltage by a quarter of a cycle. Real insulation also conducts and polarizes slightly, so a small component of current appears in phase with the voltage, and the angle by which the total current falls short of a quarter cycle is the loss angle. Its tangent is tan delta. New, dry, uniform insulation has a very low and stable loss. Water trees, moisture, contamination, thermal aging and partial discharge activity increase the loss, and they typically make it vary with voltage, because the degraded regions behave differently as the stress on them increases. Three indicators therefore carry information. The mean value reflects the overall level of loss. The stability, meaning how much the reading varies during the measurement, reflects how uniform the aging is. The change with voltage, often called tip-up, reflects how strongly the loss depends on the applied stress. Temperature, the frequency used and the cable's accessories also affect the result, which is why they are controlled and recorded.
The cable is de-energized, isolated from every source and grounded under the owner's procedures, and both terminations are cleaned and disconnected from equipment so that only the cable segment is measured. Surface leakage over terminations can be mistaken for loss, so the terminations are dry and, where necessary, guarded. The test set is connected to the conductor at one end, with the shield and the far end arranged according to the cable construction, and every other cable or circuit that would share the test voltage is disconnected. Because the cable is a large capacitor, the source must be sized for its length, and the instrument is confirmed to be in calibration. The voltage is raised in steps from a low level up to the final level the standard permits for that cable, with a timed hold and a series of measurements at each step. The operator watches both the numbers and any sign of a developing breakdown, and stops if the cable misbehaves. After the test the cable is discharged and grounded for as long as its capacitance requires. Cable temperature, ambient temperature and the length and type of the cable are recorded, and tests on the same circuit are repeated with the same settings so that results can be trended.
The record describes the cable and its accessories as far as they are known, the conditions of the test including the source, the voltage levels used and the instrument with its calibration status, and the temperature of the cable and its surroundings. The results are the loss measurement at each level, how stable it was, and anything unusual seen during the test. The state of the terminations and any companion tests are referenced as well. These details matter because a loss reading depends on age, temperature and test setup, so a later reader comparing it with a new result must be able to tell a real change in the cable from a change in how it was tested.
No single number decides whether a cable is serviceable, and this page does not give one. Published criteria take several shapes that are used together. The first is the level of the mean tan delta at defined voltage steps, judged against ranges that depend on the cable insulation type and, in some guidance, on the age of the cable. The second is the stability of the reading while voltage is held, since scatter points to localized degradation. The third is the rise of tan delta between voltage steps, where a steep climb indicates deterioration even if the mean level is moderate. The fourth is comparison with other cables of the same type and installation, and the fifth is trending against earlier tests of the same circuit. Because each indicator can lead to a different conclusion, results are usually combined into a condition category that guides action: leave in service, retest sooner, investigate further or replace. ANSI/NETA ATS-2025 and ANSI/NETA MTS-2023 address dissipation factor testing of cable; refer to your own licensed copies for the criteria, and to the cable and accessory manufacturers for their data.
At acceptance the test is applied to new cable and its terminations and splices as a check on workmanship and a baseline for later comparison, and the expected values are those of healthy new insulation. In maintenance the test is applied to cables that have been in service for years, and the interesting question is the trend and the ranking of one circuit against others. The type of insulation changes the interpretation: paper-insulated lead-covered cable, cross-linked polyethylene, ethylene propylene rubber and water-tree-retardant formulations all have their own typical loss behavior. Cables with many splices or older accessory types may show a mean value influenced by the accessories themselves. Long cables need a larger source, and cables of different lengths in a single feeder are often tested separately to avoid averaging a bad section against good ones. The test is frequently combined with partial discharge testing, because the two show different aspects of the same aging.
The equipment classes and standard sections that call for this test, and the UltraDb forms that carry it.
It describes how much energy the insulation dissipates under an alternating voltage, which rises as the insulation degrades through water trees, moisture, contamination or thermal aging. Read with its stability and its change with voltage, it gives a picture of overall condition rather than a single defect.
A long cable is a large capacitor, and charging it at power frequency requires a very large source. A very low frequency source charges the same capacitance with far less current and equipment, which makes a field test practical, while still stressing the insulation in a way that reveals loss.
When carried out within the voltage levels given by the governing guidance it is considered non-destructive, and it is chosen over withstand testing for aged cable for this reason. The operator should still watch for signs of breakdown and stop the test if the cable shows unusual behavior.
Not by itself. The result characterizes the whole segment measured. If the readings are poor, a locating method such as partial discharge mapping or time domain reflectometry, or testing of shorter sections and individual accessories, is used to find where the problem sits.
UltraDb references ASTM, ANSI, IEEE, NETA, NFPA, NEMA, FAA, TIA, and IEC standards descriptively, to identify the published methods its forms are built from. UltraDb and Verlecta are not affiliated with, endorsed by, certified by, or licensed by NETA, ASTM, ANSI, IEEE, NFPA, NEMA, FAA, TIA, IEC, or Megger. Standard names and section numbers are the property of their respective organizations. Always test to the edition your contract specifies. Where a standard is sold through the ANSI webstore, Verlecta may earn a commission on a purchase made through our link, at no extra cost to you. A commission never decides which standards we cite or how a form is built.