Dielectric Withstanding Voltage Testing for PCB Insulation
Withstanding voltage testing answers a simple question that no continuity test can: does the insulation between two conductors survive the voltage the product will actually see. The test applies a defined high voltage for a defined time and watches for breakdown or excess leakage. For boards used in mains powered, industrial or medical equipment, it is often the single most important safety check performed on the finished assembly, and at gopcb it is planned from the first drawing revision.
What Withstanding Voltage Testing Proves
The test stresses the dielectric between isolated nets and looks for a failure of that barrier. It confirms that the spacing is adequate, that the laminate is free of voids and contamination, and that no conductive debris bridges the gap. It also detects marginal insulation that would pass a low voltage resistance check but fail in the field.
Because the test applies a voltage far above the working level, it accelerates any weakness that already exists. A board that survives the full test has demonstrated a safety margin. That margin is what makes the test acceptable to safety agencies and why it appears in standards for information technology equipment, medical devices and industrial controls.
Dielectric Breakdown and How It Happens
Dielectric breakdown occurs when the electric field across an insulator exceeds the strength the material can sustain. In a circuit board that failure can run through the laminate itself, along the surface between conductors, or through a defect such as a void or an embedded particle. Each path has a different threshold.
Breakdown is rarely instantaneous in a good design; it usually begins at a weak point and propagates. Contamination, moisture, a scratch in the solder mask or a sharp copper point that concentrates the field can all begin the process. Cleanliness and smooth conductor geometry are therefore as important as the nominal spacing, and material data such as laminate material properties sets the baseline that the layout has to protect.
Test Voltage, Ramp Rate and Dwell Time
Three parameters define the test: the voltage level, the rate at which it is applied and the time it is held. The level is usually derived from the working voltage plus a safety multiple, in the range of five hundred volts to several kilovolts for mains connected equipment. Applying a higher voltage than necessary adds stress without adding information.
Ramp rate matters because a sudden step creates a capacitive current surge that can be mistaken for leakage. A controlled ramp of a few hundred volts per second gives the insulation time to settle. Dwell time is typically one minute for a type test and a few seconds for production testing, with the same pass criteria applied in both cases.
Setting Up a Hipot Test on a Board
A hipot test needs a fixture that contacts every net that must be stressed and grounds everything that should be held at reference potential. On a bare board the fixture is usually a bed of nails; on an assembly it may be a purpose built nest that isolates the connector pins involved in the safety barrier. Poor fixture design is a common source of false failures.
The reference connection has to be solid. If the ground side of the barrier is floating, the measured leakage is meaningless and the test may damage the board. Verifying the fixture before each production run, and confirming it against a test coupon with known behaviour, keeps the setup honest.
Creepage and Clearance Design Rules
Creepage is the distance along the surface of the board between two conductors, and clearance is the distance through the air. Both are specified by safety standards as a function of working voltage, pollution degree and material group. They are not the same number, and a design that satisfies one may violate the other.
Surface conditions drive creepage, which is why contamination and moisture matter so much. A slot or a routed barrier increases creepage without consuming much board area, and a groove between conductors is a common compliance measure. Designers should check these dimensions in the layout tool rather than trusting a visual estimate, and carry the required values forward in the design release checklist so that they survive later revisions.
Insulation Resistance Versus Withstanding Voltage
The two tests are complementary. Insulation resistance applies a modest voltage and measures how well the dielectric resists a small current, expressed in megohms or gigohms. Withstanding voltage applies a large voltage briefly and looks for actual breakdown. Resistance is a quality indicator; withstanding voltage is a safety proof.
A board can pass the resistance test and still fail the voltage test if the insulation has a defect that only triggers at high field. It can also fail resistance while surviving the voltage test, for example when surface contamination creates a leakage path that does not break down. Both results belong in the same record, and neither replaces the other.
Test Limits, Leakage and Pass Criteria
Pass criteria are written as a maximum leakage current, often expressed in milliamperes or microamperes, combined with the absence of breakdown. The leakage limit has to account for the capacitance of the board and the fixture, because a large board draws a real current even when its insulation is perfect. Setting the limit without measuring a known good board is a common mistake.
Leakage also changes with temperature and humidity. A board tested immediately after a humid process may read higher than the same board tested dry. Understanding that variation prevents a good product from being rejected, and it also prevents a genuine defect from being excused as normal variation.
Safety, Equipment and Fixture Design
High voltage testing is dangerous, and the equipment must be treated as such. Interlocks on the fixture enclosure, a discharge circuit for the board capacitance, and an operator who cannot touch the sample during the test are all basic requirements. Test leads should be rated for the full voltage and should be inspected regularly.
Equipment calibration matters as much as the fixture. A meter that reads low will pass boards that should fail, and a meter that reads high will create false alarms. Verifying the tester against a known load before each shift, and logging the result, turns the test into evidence rather than an opinion. Our notes on judging PCB quality describe how such records support an audit.
Recording Results and Handling Failures
Every voltage test should be recorded with the sample identifier, the test voltage, the ramp and dwell settings, the measured leakage and the result. When a failure occurs, the record shows whether the cause was the board, the fixture or the setup, and it separates a single event from a trend.
Failed boards should be examined rather than discarded. A burn mark or a carbonised path indicates a real dielectric weakness, while a repeat failure at the same net in a lot points to a design or process issue. Feeding that information back to the layout or the plating line is what prevents the next lot from repeating the problem.
FAQ
What voltage should a withstanding voltage test use? The level is set by the applicable safety standard and usually equals the working voltage multiplied by a factor of two to four, with a minimum of five hundred volts. The correct value depends on the insulation class, the pollution degree and whether the test is a type test or a routine production check.
How long should the test voltage be applied? Production testing usually holds the voltage for one to five seconds, while a type test holds it for sixty seconds. The shorter production dwell is accepted because it still reveals defects, and it keeps cycle time reasonable on a high volume line.
Can a hipot test damage a good board? A correctly specified test does not damage sound insulation, but repeated high voltage stress does age the dielectric slightly. Boards should not be tested more often than necessary, and the voltage should not be raised above the standard value simply to feel more confident.
