VLF cable testing — why a medium-voltage cable is tested at 0.1 Hz instead of power frequency after installation or repair

The guide on partial-discharge (PD) and tan-delta measurement covers a diagnostic measurement that assesses the condition of a medium-voltage cable without deliberately stressing the insulation to its limit. This article covers a different kind of test on the same cable: the VLF cable test (very low frequency), a withstand test performed after installation, repair, or a cable termination or joint to confirm that the cable can withstand an elevated voltage for a set duration without breakdown. The leading international guideline for this is IEEE 400.2, which covers VLF testing and diagnostics for extruded medium-voltage cables.

Why not simply test at power frequency

The most obvious way to test a cable would be to apply an elevated voltage at the normal power frequency (50 Hz) — comparable to normal operation, but at a higher level. The problem is the capacitive load: a long medium-voltage cable has significant capacitance, and the current needed to charge and discharge that capacitance at power frequency (the reactive charging current) increases proportionally with frequency. For a sufficiently long cable, a test transformer at 50 Hz would need an impractically large and heavy power rating just to supply this charging current — unsuitable for a portable, field-deployable test set.

The principle: the same voltage, a much lower frequency

By applying the test voltage at a strongly reduced frequency — commonly 0.1 Hz, a factor of 500 lower than power frequency — the required charging current (and thus the required test power) for the same cable capacitance decreases proportionally. This enables a compact, portable test set that can still bring a cable of considerable length to a relevant test voltage. At this low frequency, the cable's insulation is still stressed to a comparable electric field strength as at power frequency — it is the voltage, not the frequency itself, that stresses the insulation material.

Common waveforms

IEEE 400.2 recognizes several waveform types for the VLF test voltage:

Sinusoidal VLF: a smooth, sinusoidal voltage at 0.1 Hz — closest to the shape of normal operating voltage, but generally requires the heaviest test equipment. Traditionally applied to cables up to roughly 69 kV; the 2024 edition broadened the guide's scope to cables rated 5 kV and above, removing the earlier fixed upper limit.
Cosine-rectangular VLF: a waveform that switches rapidly between two polarities and maintains a nearly constant voltage in between, allowing the test set to get by with considerably less power than with a purely sinusoidal waveform.
Damped AC (DAC): a resonant discharge that produces a short-duration, decaying sinusoidal waveform, often combined with a simultaneous PD measurement.

Test voltage and duration

A VLF withstand test is typically performed at a test voltage on the order of 2.5 to 3 times the cable's rated phase-to-ground voltage, for a period of 15 to 60 minutes (with 30 minutes as a common default) — long enough to bring a marginal defect that arose during installation or repair (damaged insulation, a contamination, an incorrectly assembled joint or termination) to breakdown before the cable is put into service, without unnecessarily stressing the sound insulation of a correctly installed cable.

Note: the exact test voltage, duration, and waveform follow from the applicable standard, the voltage level and insulation type of the specific cable, and whether it concerns a test after new installation or after repair (for which a lower test voltage generally applies than for new installation); this article covers the principle, not a ready-made test table for every cable.

Practical relevance

After repairing a medium-voltage cable (for example after locating and repairing a cable fault, see the guide on fault location with the Murray bridge), the newly installed joint or termination must always be verified with a VLF withstand test before the cable is returned to service — a repair that looks visually correct can still contain a hidden assembly defect that only comes to light under an elevated test voltage.

Common mistakes

Returning a cable to service after repair without a withstand test — a visually correct-looking joint or termination can contain a hidden assembly defect that only breaks down under elevated voltage.
Applying the test voltage or duration of a new-installation standard to a repair situation, or vice versa — both situations generally have different, purpose-specific test levels.
Confusing VLF withstand testing with PD or tan-delta diagnostics — a withstand test demonstrates that the insulation survives a given voltage for a given duration, but, unlike PD or tan-delta measurements, does not provide a quantitative trend of insulation condition over time.
Not accounting for the already-connected accessories (joints, terminations) of the cable during the VLF test — the entire chain of cable and accessories is tested, so a weak point in an accessory can still cause an otherwise sound cable to fail the test.