- Test voltage
- 0 to 500 V, continuously adjustable
- Leakage ranges
- 0.01 to 3 µA, 3 µA to 3 mA, 3 mA to 30 mA
- Reading error
- ±1% ±2 digits
- Charge time
- 0 to 99 s, ±2%
- Discharge time
- 0 to 9 s, ±2%
- Maximum charging current
- At least 200 mA
- Maximum test current
- At least 30 mA
- Terminals
- Two-terminal, shielded, with an outer shield ground
- Mains
- 220 V ±10%, 50 or 60 Hz ±5%
- Body
- 330 × 320 × 120 mm, about 5 kg
A leaking electrolytic capacitor rarely fails outright. It drifts, heats, and dries out over a service life that is shorter than the one on the datasheet, and the only warning is a leakage current that has climbed past its limit. Measuring that current is simple in principle and awkward in practice, because it is a microamp sitting on top of a charging current many thousand times larger.
Product overview
This is a single-model bench instrument for that measurement. It supplies an adjustable direct voltage up to 500 V, holds the capacitor at that voltage for a set time, measures the residual current and then discharges the part before the fixture is opened.
The front panel carries a digital display, the voltage and time settings, and the two shielded test terminals. The case is 330 mm wide at about 5 kg, and a high-voltage warning is fitted to the body. The instrument is supplied with a mains lead and a test fixture for the capacitor package.
How the leakage tester works
The instrument charges the capacitor through a current-limited source until the terminal voltage reaches the set value, then holds that voltage while the current falls. What is displayed after the hold is the leakage current: the small conduction through the dielectric and the seal that continues once the plates are fully charged.
The hold time is the part of the procedure that decides the reading. A capacitor that has just been charged is still absorbing, and its current is falling steeply; the same capacitor a minute later sits on a much flatter curve. Setting the charge time between 0 and 99 s fixes a point on that curve, so that two capacitors, or two production lots, are compared at the same place. The discharge time, up to 9 s, is set for the same reason — to bring the part down predictably before it is handled.
Three current ranges cover 0.01 µA to 30 mA, and the reading error is ±1% ±2 digits. The over-limit indication, at ±3%, is what turns the reading into a pass or fail without the operator comparing numbers.
How to select the right leakage tester
There is one model here, so selection is about matching the procedure rather than choosing between variants. The first question is the test voltage: it has to reach the rated voltage of the parts, and 500 V covers the common 400 V and 450 V power-supply cans as well as low-voltage parts.
The second is capacitance, because it sets the charging time. A small part reaches its test voltage in a second; a large can draws the full charging current for much longer, and if the charge time is shorter than the time needed to reach the voltage, the instrument is still measuring charging current when the reading is taken.
The third is the leakage limit itself. A limit in the low microamps needs the lowest range and a stable fixture; a limit in the milliamps is easier to measure but more likely to be affected by the part heating during the test. The fourth is the fixture: terminal spacing, can diameter and whether the part is tested before or after the seal is fitted decide the jig more than the instrument does.
Send the capacitance, the rated voltage, the leakage limit and the hold time in the procedure, and the fixture can be specified with the instrument.
Measurement limitations and selection considerations
The measurement is made at direct voltage, so it describes the dielectric under a steady field and not the behaviour of the capacitor in a circuit. Equivalent series resistance, capacitance tolerance and impedance at frequency are separate measurements, and a capacitor that passes a leakage test can still be a poor part for a switching supply.
Temperature is the dominant influence on leakage current. A part measured cold and a part measured warm will not agree, and the difference can be larger than the tolerance band. Testing at a controlled ambient, or at least at a consistent one, matters more than the instrument’s own accuracy figure.
The hold time also has to be long enough for the reading to be meaningful on a large can, which is the practical reason the charge timer goes to 99 s. Where the procedure calls for a longer soak than that, the instrument has to be used with an external timing step rather than stretched beyond its setting.
Finally, the reading includes the leakage of the fixture and of the surface between the terminals. A dirty fixture at 500 V can leak more than the capacitor being tested, which is why the terminals are shielded and why a zero check with no part fitted belongs in the procedure.
Every value below is taken from source documentation. Where a parameter is not stated in the available material it is marked as information required rather than estimated.
