VA Technology 
IEEE 400.2-2024, published in September 2024, was the first revision of the field-testing guide for shielded power cable since 2013, and it keeps tan delta inside the diagnostic set that runs on the same low-frequency excitation as a withstand test. That is the point of VLF tan delta testing: the withstand ends in pass or fail, while the loss measurement beside it says how much margin remains.
What tan delta actually measures
An insulated cable is a capacitor, but an imperfect one: a small part of the current driven into the insulation is dissipated as heat rather than stored and returned. The loss angle δ is the amount by which the cable current departs from a perfect 90° lead, and tan δ is its tangent — the ratio of resistive to capacitive current, and so a direct measure of how lossy the insulation has become. Moisture, contamination and ageing products all raise it.
Tan δ is a ratio and not an absolute current, so it survives the move to very low frequency — but the loss mechanisms do not behave identically at 0.1 Hz and at power frequency, so readings are not converted between the two. VLF tan delta testing reads against the same cable’s own history, and against how the value responds as the test voltage is stepped.
Why a single number is not the answer
A tan delta taken at one voltage is a single point. The diagnosis comes from stepping the voltage and watching the loss respond, because defects scale differently.
| Pattern across the voltage steps | What it points to | How to treat it |
|---|---|---|
| Low and flat | Insulation behaving as designed | Record it as this cable’s baseline |
| Elevated but flat | Bulk ageing or moisture spread through the insulation | Trend it; check the sheath and the joints |
| Rising with voltage (tip-up) | Water trees, voids, localised non-linear loss | Prioritise; plan repair or replacement |
| Unstable between steps | Active discharge, or a termination problem | Repeat in isolation before deciding |
The shapes most often confused are the first and the third. A loss that is high but flat is a property of the whole insulation volume; a loss that grows with voltage is concentrated somewhere, and that somewhere is where the cable will fail.
Doing it in the field
The instrument has to hold a stable sinusoidal output while resolving a loss that is a small fraction of the current it drives. The two tan-delta instruments in the VA-TEK family — the VLF-34TD and the VLF-45TD — carry integrated tan delta over a 5 nF to 10 µF load range, with an accuracy of 1.0 × 10⁻⁴ and a resolution of 1.0 × 10⁻⁵, at four selectable frequencies between 0.1 Hz and 0.01 Hz.
Frequency selection is a tool. A lower test frequency lets the set drive a larger capacitance: the VLF-34TD holds 1.0 µF at 34 kV peak at 0.1 Hz and 5 µF at 0.02 Hz, so the slower setting reaches further along a long run. Across two frequencies, a loss that holds steady behaves like a bulk property; a strong frequency dependence points elsewhere.
Frequently asked questions
Is tan delta a pass or fail test?
Not from one reading. VLF tan delta testing is a comparison against the same cable’s earlier result and against the shape of the curve in front of you. A cable reading the same as it did three years ago sits differently from one whose loss has doubled, even when both numbers look modest.
What tan delta value is acceptable?
There is no single universal threshold. Limits vary with cable type, age, construction and the owner’s own policy, and a number borrowed from another network is not a diagnosis. Two things travel better than a borrowed limit: the direction of travel over time, and whether the loss rises with test voltage.
Does every VLF test set measure tan delta?
No. A withstand set only has to produce a stable high voltage and measure current; resolving a loss tangent near 10⁻⁴ needs a different front end and a defined load range. In the VA-TEK family the two tan-delta instruments carry it, the VLF-34TD for 5–15 kV cable and the VLF-45TD for 5–25 kV cable. The VLF-80 and VLF-90 are withstand machines.
VA-TEK builds the VLF hipot family for acceptance and maintenance work on shielded cable, from the high-voltage withstand units down to the diagnostic machines that add the loss measurement. Where a cable class needs a voltage the family cannot reach, we say so; the derivation of those test levels sits in our VLF versus DC article. Send us the cable class and route length and we will match the set to the job.
Sources
- IEEE 400.2-2024, IEEE Guide for Field Testing of Shielded Power Cable Systems Using Very Low Frequency (VLF): published September 2024, supersedes IEEE 400.2-2013; tangent delta testing listed among the covered diagnostic methods.
- IEC 60502-2, Power cables with extruded insulation and their accessories for rated voltages from 1 kV to 30 kV: in-situ AC testing.
- GB/T 3048.8-2025, Test methods for electrical properties of electric cables and wires — Part 8: AC voltage test: effective 1 May 2026; VLF and oscillating-wave annexes.
- VA-TEK product data, VLF-34TD and VLF-45TD: tan delta accuracy 1.0 × 10⁻⁴, resolution 1.0 × 10⁻⁵, load range 5 nF to 10 µF, output frequencies 0.1 / 0.05 / 0.02 / 0.01 Hz, load ratings 1.0 µF at 34 kV at 0.1 Hz and 5 µF at 0.02 Hz (vendor data).
