Key takeaways
- Resistance catches wire size, turn count and connection faults, but only when corrected to a reference temperature.
- Inductance without a stated frequency, bias and fixture is not a specification.
- Hi-pot proves insulation integrity, and its voltage, ramp and leakage limit belong in the test procedure.
- Turns ratio and polarity checks catch wrong connections that a resistance reading can miss.
A test sequence, not a test list
A coil test plan is a sequence. The first test should be the one that is safe to run on an untested part and the one that catches the most common fault. Resistance comes first because it is quick, it needs no high voltage and it shows an open circuit or a wrong wire size before any other test is applied.
Inductance comes next, with the core and the fixture in the production condition. The hi-pot test follows, because it stresses the insulation and should not be applied to a part that is already known to be shorted. Turns ratio and polarity complete the set where the coil has more than one winding.
The sequence also matters for the product. A hi-pot test can leave a charge on the winding, so the fixture needs a discharge step. A test after potting may read differently from a test before potting, and both readings may be useful. State the stage at which each test is made.
Resistance
Resistance is the simplest check on the conductor. It responds to the wire diameter, the turn count, the mean length per turn and the quality of the termination. A reading that is high can mean a wrong wire, a broken strand, a poor solder joint or a partly open connection.
Use a four-wire measurement for a low resistance, so the lead and contact resistance do not dominate the result. Record the temperature and correct the reading to a reference, usually 20 C. Copper changes by about 0.393 per cent per kelvin, so a coil measured hot will read higher than the same coil cold. Without the correction, a good coil can look out of limit.
For a multi-winding coil, measure each winding and the balance between them. A large imbalance points to a wrong turn count or a short between turns. A small imbalance may be normal, depending on the design and the winding sequence.
Inductance
Inductance is measured with an LCR meter or a dedicated coil tester at a stated frequency. The reading depends on the frequency, the test voltage, the DC bias and the fixture. A value quoted without those conditions cannot be compared with a later measurement.
For a gapped core, the inductance at low signal may be much higher than the value at the operating current. If the application carries a DC bias, measure at that bias. If the core saturates, the inductance will fall, and a small-signal test will not show it.
The quality factor, Q, is the ratio of reactance to resistance at the test frequency. It is a useful indicator of the winding loss and the core loss, but it is not a substitute for a resistance measurement. A coil can have a good Q and a poor DC resistance if the wire is thin and the turns are many.
Hi-pot
The hi-pot test applies a high voltage between the winding and the core, the frame or another winding, and measures the leakage current. It proves that the insulation can withstand the stress for the test duration. It does not prove the creepage distance, the clearance or the long-term ageing.
State the test voltage, the waveform, the ramp, the dwell and the leakage limit. An AC test stresses the insulation differently from a DC test, and the two are not interchangeable. A DC test can charge the capacitance of the winding, so the ramp and the discharge need care.
The test voltage should be high enough to find a defect and low enough not to damage a good part. Repeating a hi-pot test at the same stress can degrade the insulation over time, so the number of applications belongs in the procedure. The fixture must be guarded and interlocked, because the test voltage is lethal.
Hi-pot is a pass or fail, not a measurement of margin
A part that passes at the test voltage has survived that stress. It does not tell you how close it was to breakdown, and it does not replace a creepage check.
Turns ratio and polarity
A turns ratio test compares the voltage induced in one winding with the voltage applied to another. It catches a wrong turn count, a shorted turn, a wrong bobbin or a winding connected to the wrong terminal. The ratio should be close to the design value, within the tolerance set by the winding and the measurement.
Polarity is the phase relationship between windings. A reversed winding can still have the correct ratio and the correct resistance, but it will not work in the circuit. Mark the phase dots on the drawing and check them on the tester. A polarity error is easy to introduce when the start and finish leads are not clearly identified.
An impulse or surge test can find a shorted turn that a resistance measurement misses. It compares the decay waveform of the test winding with a known good winding. The method is fast and sensitive, but it needs a reference and a trained operator to interpret the trace.
Tests, faults and limits
The table maps each test to the fault it finds and the limit it needs. The limits are placeholders until the drawing and the standard set them; the important point is that each test has a stated condition and a stated acceptance rule.
| Test | What it finds | Limit to state | When to run |
|---|---|---|---|
| Resistance | Wire size, turn count, open or poor joint | Value at 20 C with tolerance | Every coil, before potting |
| Inductance | Turns, core, gap and saturation | Value at stated frequency and bias | Every coil or audit, at the same stage |
| Q or loss | Winding and core loss | Minimum at the test frequency | Audit or type test |
| Hi-pot | Insulation breakdown and leakage | Voltage, ramp, dwell and leakage limit | Every coil, before and after potting as specified |
| Turns ratio | Wrong turns or wrong winding | Ratio and tolerance | Every multi-winding coil |
| Polarity | Reversed winding or lead | Phase relation against the drawing | Every multi-winding coil |
| Insulation resistance | Moisture and contamination | Minimum at stated voltage | Audit, after humidity or potting |
The order is part of the control plan. A test run at the wrong stage can pass a fault that appears later.
Records and disposition
Record the test conditions with the result: the instrument, the calibration date, the fixture, the temperature, the frequency, the bias and the operator. A resistance value without a temperature, or an inductance without a frequency, cannot be compared with the next lot.
Use a known good sample to check the fixture at the start of each shift. A drifting fixture can pass bad parts or reject good ones, and the trend is easier to see on the sample than on the production data. Keep the sample with the tester.
When a coil fails, keep it and record the failure mode. An open circuit, a shorted turn, a low inductance and a hi-pot failure point to different process steps. The failure analysis should name the step and the change, not just the test that rejected the part.
Frequently asked questions
Why does coil resistance change with temperature?
Copper resistance rises with temperature, by about 0.393 per cent per kelvin near room temperature. A coil measured hot reads higher than the same coil cold. Correct the reading to a reference temperature, usually 20 C, before comparing it with the limit.
What does a hi-pot test prove?
It proves that the insulation survived the applied voltage for the stated time without exceeding the leakage limit. It does not prove the creepage distance, the clearance, the long-term ageing or the impulse withstand. Those need their own checks.
Can a turns ratio test find a shorted turn?
It can show a ratio error if the short changes the effective turns. A single shorted turn may produce a small change that is hard to see. An impulse or surge test is more sensitive to a shorted turn and is often used alongside the ratio test.
Should every coil be hi-pot tested?
That depends on the standard, the application and the risk. Many programmes test every coil because the test is fast and a breakdown is a safety issue. The procedure should state the voltage, the ramp, the dwell, the leakage limit and how many times the test may be applied.
Sources and further reading
- IEC 61558-1, Safety of power transformers, power supplies, reactors and similar products
- IEC 60076-1, Power transformers, general
- Qumek coil test procedure and fixture validation notes
Qumek Engineering Desk
Engineering team
The engineering team behind Qumek drawings, DFM notes, tolerances and inspection standards. Every guide is reviewed against the drawings and processes used on live programmes.
Writing a coil test procedure?
Send the winding data, the insulation requirement and the production stage. The engineering desk will reply with a test sequence and the limits to state.
Send the winding data