5.3 Overpotential vs IR & Cable Hipot Basics

IR Test vs Overpotential (Hipot) Test

These two tests are routinely confused because both apply DC or AC voltage to insulation. They differ in purpose, magnitude, duration, and pass criterion.

An IR test tells you how good the insulation is; a hipot tells you whether it survived. An IR test is a measurement; an overpotential test is a proof test. The hipot applies voltage well above the operating level — for medium-voltage cable, often 3 to 4× the rated line-to-ground voltage — to stress the dielectric the way a transient or switching surge might. If the insulation withstands for the hold time without flashover or puncture, it passes; if it arcs over, it fails, and the failure point is often physically damaged.

Why Overpotential Stresses Insulation Differently

The higher voltage drives the dielectric stress above its normal operating field. Defects that are invisible at operating voltage — voids, contamination tracks, mechanical damage — ionize and break down under overpotential. That is the test's value (it finds weak insulation) and its risk (a marginal cable that would have run for years may be punctured by the test). For this reason, IR and PI/DAR are always run before a hipot: a low IR reading means the insulation is wet or dirty, and hipot-ing wet insulation is likely to damage it rather than prove it. Sequence: clean and dry → IR → PI/DAR → hipot. Never hipot wet or contaminated insulation.

Cable Hipot Basics: DC vs AC vs VLF

Cable hipot comes in three flavors, each with a different physics and a different appropriate cable type.

DC Hipot

DC hipot applies a high DC voltage (tens of kV for MV cable) and measures leakage current. Because DC draws only the tiny conduction and absorption currents, the test set is small and the current is a sensitive diagnostic. DC hipot is the traditional acceptance method for paper-insulated lead-covered (PILC) cable and is still permitted there. However, since the early 1990s DC hipot has fallen out of favor for extruded dielectric cables (XLPE, EPR): evidence shows high DC voltage can create space charge in the extruded dielectric and cause latent damage that leads to later in-service failure. For XLPE/EPR, NETA and IEEE 400.2 now prefer VLF and partial-discharge methods.

AC Hipot (Power Frequency or 20 to 300 Hz)

AC hipot stresses insulation more uniformly than DC because the field reverses each half-cycle, mimicking service conditions. It is used for apparatus (switchgear, bus, transformers) and for cable where a resonant test set (20 to 300 Hz) is available. AC test sets are larger because the cable capacitance draws significant charging current at power frequency.

VLF (0.1 Hz) Hipot

Very Low Frequency hipot applies AC at 0.1 Hz, which charges the cable capacitance 500× more slowly than 60 Hz and so needs a fraction of the current. Per IEEE 400.2-2024, VLF is the preferred withstand test for extruded-dielectric MV cable. The VLF withstand voltage is roughly 1.7 × U₀ (conductor-to-ground voltage) for acceptance, held for a defined time (commonly 15 minutes for acceptance, shorter for maintenance). VLF tan-delta and VLF partial-discharge diagnostics are also defined in IEEE 400.2 and give more condition information than a bare withstand test.

NETA MTS Cable Acceptance Voltages

NETA MTS Tables 100.6.1 through 100.6.5 give the specific acceptance and maintenance voltages by cable type and voltage class (DC, VLF, DAC, and AC 20 to 300 Hz). For a 15 kV class shielded cable, DC acceptance is on the order of 45 to 55 kV DC, with maintenance values roughly 60 to 65% of acceptance. For VLF, follow IEEE 400.2-2024 and NETA Table 100.6.3 — the VLF voltage is lower than the DC equivalent because the AC field reverses every 5 seconds and stresses the dielectric more effectively. Exact values are in the NETA tables and the manufacturer's published data — always confirm against the current edition of ANSI/NETA ATS-2025 or MTS and do not rely on memory for the exact kilovolts.

Cable Step-Voltage DC Test

The step-voltage test is a diagnostic version of DC hipot. Instead of a single elevated voltage, the voltage is raised in equal steps — for example 5 kV, 10 kV, 15 kV, 20 kV, 25 kV — holding each step long enough for leakage current to settle, and plotting leakage current versus voltage.

What a Healthy Cable Looks Like

Healthy insulation is ohmic over its working range: leakage current rises linearly with applied voltage, so a plot of current vs voltage is a straight line through the origin and resistance is constant at every step. That linear, stable behavior at each step is the pass signature.

What a Rising (Non-Linear) Leakage Current Means

A disproportionate increase in leakage current at a higher step — a knee-point where current jumps more than proportionally — indicates the onset of insulation breakdown or contamination. The dielectric is no longer ohmic; carbonized tracks, moisture, or a void are conducting at the higher field. The voltage at which the knee occurs is the breakdown onset voltage, and trending that knee voltage downward over years is an early warning of approaching failure.

Why Overpotential Stresses Differently Than IR

A single IR reading at 1,000 or 5,000 V probes the dielectric at a low field; the step-voltage test ramps the field until the dielectric reveals its non-linearity. The IR test answers "how good is it?"; the step-voltage test answers "at what voltage does it start to break down?" A declining IR trend across steps — resistance falling as voltage rises — is the classic degradation signature and can catch a failing cable long before a single-voltage test would.

Trap

For extruded (XLPE/EPR) cable, do not rely on DC step-voltage as the acceptance test — use VLF or partial discharge. DC step-voltage remains a useful diagnostic for PILC and for apparatus where space-charge concerns do not apply. Also remember that a hipot is a pass/fail proof test, not a trending measurement — a cable that withstands 50 kV DC for 1 minute has proven it survived, but the leakage current value itself is not a reliable trend quantity the way an IR megohm reading is.