Dielectric Strength and Hipot Testing: What the Isolation Rating Means
Dielectric strength testing is how a manufacturer proves that the isolation inside a driver actually isolates. The test applies a high voltage between the mains terminals and the output terminals and watches for breakdown or excessive leakage. It is performed on every unit in production as a routine test and on a sample as a type test, and the voltage used is far above anything the product sees in normal service. The number on the certificate, often 3.75 kV AC, is the evidence behind the isolation claim.
What the number means depends on the standard. For reinforced insulation between a mains input and a SELV output, IEC 61347-1 sets the test voltage according to the working voltage and the insulation class. A driver rated for 240 volts mains typically sees a type test of several kilovolts applied for one minute, and a production test at the same or a slightly lower level applied for one second. The pass criterion is no breakdown and a leakage current below a specified limit, measured with test equipment calibrated for the application.
AC is used because it stresses the insulation the way the mains does. A DC test would charge the capacitance of the isolation barrier and could pass a defect that AC would flash over, and it leaves the barrier charged after the test. AC testing applies a sinusoidal voltage at the power frequency, and the measured leakage includes the capacitive component, which is why the limit is set with the barrier capacitance in mind rather than as a simple resistance measurement.
The test is not free of risk to the product. Repeated hipot testing at high voltage can degrade insulation, particularly at weak points and along contaminated surfaces, and a test on a complete luminaire can damage surge protectors, capacitors and sensitive electronics connected across the barrier. For this reason the routine test is done at the driver stage, before assembly, and hipot testing of installed fixtures is not a normal maintenance activity.
Reading a test report means checking three things: the voltage, the duration and the leakage limit, and matching them against the standard the product claims. A report that lists a voltage without the standard, or a duration that does not match the type or routine test requirement, does not prove anything. The report should also identify the test points: input to output, input to earth for Class I, and output to earth where applicable.
For the PV100, the isolation between the mains input and the 24V output is what allows the output to be treated as SELV, and the hipot result is part of the safety file behind the CE marking. A buyer who needs to verify the claim should ask for the certificate and the test report rather than relying on the label. The voltage printed on the case is a summary; the report is the evidence.
The distinction between type test and routine test matters in procurement. A type test proves the design; a routine test proves the unit. A supplier who can show only a type test report has not demonstrated that every unit was tested, and a supplier who shows only a routine result has not demonstrated the design margin behind it.
Routine tests are usually performed at a lower voltage and a shorter duration to avoid stressing the product, and the values are recorded per unit. For large projects, sampling the production records is a reasonable way to confirm that the test is actually performed rather than claimed, and the sample should include units from different production dates.
Output Ripple: Where It Comes From and How It Reaches the LEDs
Inrush Current Limiting: NTC Thermistors and Active Bypass Circuits
Related Article