VA-TEK Engineering During a VLF withstand test the operator watches one number above all others: the current. It is easy to read that number as a health indicator on its own — low is good, rising is bad — but most of the current flowing into a cable at 0.1 Hz is not leakage at all. It is the capacitive charging current, and it is a property of the cable’s geometry far more than of its condition. Knowing which part of the trace responds to insulation and which part responds to the test setup is what turns a pass/fail withstand into a diagnostic measurement.
This article explains what the current trace during a VLF test is made of, how the frequency choice changes it, and which changes during the hold actually matter.
Key takeaways
- At 0.1 Hz the current is dominated by the capacitive charging current, I = 2πfCU — it scales with frequency, cable length and test voltage, not with insulation health.
- The diagnostic component is the much smaller in-phase (resistive) current. Its trend and shape matter more than its absolute value.
- The same cable that draws a few milliamps at 0.1 Hz would draw amps at 50 Hz — this is why a VLF set is portable.
- Stored energy does not scale with frequency: a discharged cable holds the same joules at 0.1 Hz as at 50 Hz, which is why discharge and earthing procedure is unchanged.
- Record voltage, frequency, capacitance, current and temperature with every trace. A current figure without its basis cannot be trended.
1. What the current trace is actually made of
A cable under AC test voltage behaves as a capacitor with a small loss. The total current has two components at right angles: a charging current that leads the voltage by 90°, and a loss current in phase with the voltage that represents energy dissipated in the insulation. The magnitude of the charging current follows the standard capacitive relation:
Ic = 2πf · C · U
Take a 3 km, 11 kV XLPE cable with roughly 0.25 µF per kilometre, so about 0.75 µF in total, tested at 15 kV. At 0.1 Hz the charging current is about 7 mA. The same cable at 50 Hz would draw about 3.5 A. Nothing about the insulation changed between those two figures; only the frequency did. The loss current, by contrast, is a small fraction of the charging current and is the part that responds to moisture, contamination and ageing.
2. Why 0.1 Hz makes the current both manageable and readable
The frequency reduction is the reason a VLF set fits in a van. It also changes what the operator sees. At 0.1 Hz a 30-minute hold gives 180 cycles; at 0.02 Hz the same hold gives 36. Each cycle is a full polarity reversal, and the shape of the current through those reversals carries information that a single rms reading discards.
The trade-off is that the loss current is proportionally tiny, so a raw current display is a blunt instrument for insulation condition. That is the reason VLF tan delta exists: it measures the ratio of loss to charging current directly, which is far more sensitive than watching the total move. The relationship between the two measurements is set out in VLF tan delta testing.
3. Reading the trace during the hold
What matters is not the first number but how the trace behaves between voltage application and the end of the hold. The table below maps the common patterns to their likely meaning.
| Observation | Likely meaning | Response |
|---|---|---|
| Stable current, symmetrical half-cycles | Normal capacitive behaviour | Continue to full hold time; record the value with voltage and frequency |
| Slow, smooth rise through the hold | Thermal or moisture-driven loss increase | Note the slope and temperature; trend against previous tests of the same circuit |
| Rise that tracks ambient temperature | Loss current is temperature-dependent | Compare tests at similar conductor temperature, not across seasons |
| Step change in current | Partial breakdown or a developing defect | Stop the test, discharge, and locate before re-testing |
| Asymmetry between polarities | Space charge, moisture or a rectifying defect | Investigate; this is not normal sine behaviour |
| Current collapses to near zero | Loss of connection, trip, or test-set protection | Check the circuit before assuming a cable result |
Two of these patterns are routinely misread. A slow rise is often assumed to be a failing cable when it is simply the insulation warming under test voltage — which is why the temperature record matters. And a current that looks “high” compared with another circuit may only reflect a longer cable: 0.25 µF per kilometre means a 6 km circuit draws twice the charging current of a 3 km one at the same voltage and frequency.
4. What the current rating of the set actually limits
Because charging current scales with capacitance, frequency and voltage, the test set’s output current rating sets the maximum cable length it can charge at a given test voltage and frequency. This is the constraint that decides whether a longer circuit can be tested at 0.1 Hz or has to drop to 0.05 or 0.02 Hz. Dropping the frequency reduces charging current in proportion and lets the same set test a longer cable, at the cost of fewer cycles within the hold.
That calculation belongs in test planning, not in the field. The selection logic for voltage class and load is covered in choosing a VLF test set by voltage class, and the frequency choice itself in VLF test duration and frequency.
5. The current is not the safety number — the energy is
A cable charged to test voltage stores energy in its capacitance regardless of the test frequency. For the same 0.75 µF cable at 15 kV, that is roughly 84 joules — 0.5 × C × U². The low frequency lowers the current the set must deliver, not the charge waiting in the cable at the end of the test. Discharge and earth the cable after every test, discharge again after a pause because dielectric absorption can rebuild voltage, and treat the circuit as live until it is proven dead at the point of work.
6. What to record with every trace
Record test voltage and its basis (peak or rms), frequency, estimated or measured capacitance, current at the start and end of the hold, conductor or ambient temperature, cable length and route, and the waveform if the set stores one. Those fields make a result comparable with the next test of the same circuit — and comparability, not any single number, is what makes a VLF programme useful over a cable’s life.
For the method as a whole, see the practical VLF cable testing guide; for how test voltage is chosen, VLF test voltage selection; and for discharge measurement under VLF, VLF partial discharge testing. The VLF hipot test set range lists the output voltage and current ratings of each model, from 30 kV to 90 kV peak, including the TD variants with tan delta capability.
If you are sizing a set against a cable schedule, send the lengths, voltage classes and test regime to our application engineers through the contact page and they will confirm which model and frequency settings cover the work.
