VLF vs DC Cable Testing: Why 0.1 Hz Replaced DC Hipot

For shielded power cable the field proof test has moved from direct current to very low frequency alternating current. The reason is physics: DC leaves trapped space charge in extruded insulation, and that charge can turn a sound cable into a failure. Vlf cable testing reverses polarity every half cycle, so no charge builds.

Why DC testing was abandoned on extruded cable

DC still suits paper-insulated lead-covered cable, where a lapped paper dielectric does not accumulate charge the same way. Cross-linked polyethylene (XLPE) and ethylene propylene rubber (EPR) behave differently: charge migrates to defects and to the insulation-screen interfaces, and the field that builds there can exceed anything the cable sees in service. A cable that would have passed an alternating test can fail under DC, or soon after it returns to service.

Low frequency is the compromise. At 0.1 Hz the waveform still alternates, so space charge is neutralised each half cycle, while the charging current falls by a factor of about five hundred against 50 Hz, because capacitive current is proportional to frequency. That is why a portable set can hold the test voltage on a kilometre of cable, and why vlf cable testing became the default for extruded insulation.

How the test voltage is set

The figure that matters is U0, the cable’s rated phase-to-earth voltage — the system voltage divided by the square root of three, not the phase-to-phase number on the drum. IEEE 400.2 ties acceptance testing of new cable, joints and terminations to roughly 3U0, and maintenance testing of aged cable to about 2U0, because a new-cable level can fail an aged cable. The exact multiple and duration depend on cable type and test category; take the certified figure from the standard’s tables.

One boundary is worth stating. A 66 kV cable has a U0 of about 38 kV, so its acceptance level sits near 114 kV, above the 90 kV peak of the largest unit here: the VLF-80 and VLF-90 reach the maintenance level for that class, and 66 kV acceptance needs a higher-voltage host. Ratings here are peak values, never multiplied by the square root of two.

What the VLF range covers

The VLF hipot testers span six withstand units from 30 kV to 90 kV peak — the VLF-30, VLF-40, VLF-50, VLF-60, VLF-80 and VLF-90 — plus two diagnostic machines, the VLF-34TD and VLF-45TD, for 5 to 15 kV and 5 to 25 kV cable. All share the same blocks: a sinusoidal output at 0.1, 0.05, 0.02 and 0.01 Hz, 0 to 30 mA peak continuously, a load rating up to 5.5 µF at 0.01 Hz, and accuracy of ±3 % of reading.

Each unit splits into a 4 kg controller and a 20 to 55 kg high-voltage tank. The VLF-60 also covers motor and generator stator insulation, where a power-frequency set could not charge the winding capacitance. The two TD machines add tan-delta diagnostics at 1.0×10⁻⁴ accuracy with 1.0×10⁻⁵ resolution, plus DC and sheath testing with sheath fault location.

Every unit is CE marked for safety and EMC, and the design is referenced to IEEE 400.2-2024 and IEC 60502-2. To match a unit to your cable classes, contact VA-TEK or request a quote.

Residual voltage: the hazard after the test

A tested cable is a charged capacitor, and even a small set is rated for circuits storing more energy than the numbers suggest: a kilometre of 15 kV cable holds tens of joules at acceptance voltage, a 35 kV cable hundreds, enough to arc, burn or kill.

In extruded insulation the decay is not one clean exponential: dielectric absorption gives several time constants, so the voltage can partly reappear after an initial discharge, which NETA World’s Fall 2026 field note treats as the specific hazard. Ramp down, discharge through an insulated stick with current-limiting damping resistors, apply solid grounds to every phase, and measure the residual voltage before touching the circuit. The automatic internal discharge is the first step, not the last.

Frequently asked questions

Is a DC hipot test ever the right choice?

For paper-insulated lead-covered cable, yes: that dielectric does not accumulate space charge the way an extruded one does. On XLPE and EPR, DC is the test that can cause the failure it was meant to find, which is why vlf cable testing displaced it.

How long does a VLF cable test take?

The duration is set by the standard and the project specification, and runs to tens of minutes per phase rather than seconds — the low frequency is what makes vlf cable testing practical on a long, high-capacitance feeder.

Does a VLF test also locate the fault?

A withstand test answers pass or fail and nothing more. The VLF-45TD adds sheath testing with sheath fault location to IEC 60229, which turns a failed sheath into a position. The transformer-side equivalent is in the transformer electrical testing guide.

Sources

• IEEE 400.2-2024 — field testing of shielded power cable systems using VLF.
• IEC 60502-2 — extruded-insulation power cables, 1 kV to 30 kV.
• IEC 60229 — cable sheath testing and sheath fault location.
• NETA World, Fall 2026 — “Residual Voltage in Power Cables: A Little-Known Hazard in Field Testing”.
• VA-TEK product data (vendor data) — VLF output and load ratings.