Capacitor Leakage Current Tester

A capacitor that measures correctly on an LCR meter can still fail on the line, because capacitance is not the property that decides whether it survives a voltage. Leakage current is, and it is measured at the voltage the part will actually see.

Product overview

The range covers four bench instruments built around the same job: apply a direct voltage, hold it long enough for the capacitor to charge, then read the small current that keeps flowing. What separates the models is how far the voltage goes, how small a current can be resolved, and how the result is displayed.

Two of them carry a dot matrix display with a full numeric keypad and a measured insulation resistance function; the other two are narrower instruments with a fixed 200 V and 500 V selection, one under microprocessor control and one built around an analogue movement. The pair with the wider voltage range is intended for parts rated above what the smaller pair can test.

How the capacitor leakage tester works

The instrument regulates its output to the set voltage and limits the current while the capacitor charges. That current limit is the reason a large electrolytic can be brought up to test voltage without drawing a surge: the instrument holds the charging current at the set value, and the part reaches its voltage over a controlled interval rather than instantly.

Once the charging time has elapsed, the current that continues to flow is the leakage current. It is a very small quantity, so the measurement is made at the low end of the instrument’s range and quoted with a fixed offset alongside the percentage term, because at the bottom of the range the offset dominates.

Insulation resistance is calculated from the same current and the applied voltage. On the models that provide it, the resistance and the leakage current are two views of one measurement, which is why the limits for both can be set on the same instrument and judged by the same comparator.

The comparator is the part that turns the reading into a decision. Limits are entered for the parameter being judged, and the instrument reports a pass, a fail or an alarm, which is what allows the test to be run by an operator who is not reading the numbers.

How to select the right capacitor leakage tester

Begin with the test voltage. A part rated above the upper limit of a model cannot be tested on it at its rated voltage, and testing below the rating proves less than the drawing asks for. The two models with the extended range exist for that reason alone.

Then look at the smallest leakage current the inspection has to resolve. Small film and ceramic parts leak in the nanoamp region, where the resolution of the instrument and the fixed term in its accuracy statement decide whether a limit can be set at all. Large electrolytics leak in the milliamp region, where that term is irrelevant and the higher current range matters instead.

Then consider the charging time the part needs. A large can takes longer to settle than a small one, and the timer range has to cover the interval the test procedure calls for. The models differ here as well as in voltage.

Finally, decide how the result leaves the bench. A stand-alone test uses the comparator and the display; a station that logs results or hands the part to a handler needs the serial interface and the handler lines, and the optional bus interface if the station is already built around one.

Send the capacitance, the voltage rating, the leakage limit and the charging time from the inspection sheet, and the model and fixture can be quoted against them.

Measurement limitations and selection considerations

The measurement is destructive in one direction: a capacitor taken above its rated voltage can be damaged in a way that only shows later. The test voltage has to come from the part specification, not from what the instrument can produce.

Leakage current depends on temperature and on the time since the voltage was applied. A reading taken before the dielectric has settled is higher than the steady value, which is why the charging interval is part of the test procedure and why two laboratories using different intervals will not agree on the same part.

The instrument measures the total current flowing into the capacitor, so a fixture that leaks will add to the reading. Cables, contacts and any surface contamination on the jig are in parallel with the part under test, and at the nanoamp level that path can be larger than the leakage being measured. Cleaning and guarding the fixture is part of making the measurement, not housekeeping.

Finally, the comparator only judges what it is set to judge. A part that passes a leakage limit has passed that limit at that voltage and temperature; nothing about its life, its equivalent series resistance or its behaviour in a switching circuit follows from the result.