Insulation and dielectric
A dielectric withstand test, often called a high-potential or overpotential test, applies a voltage well above normal operating level for a timed interval to prove that insulation can hold without breaking down. It is a pass or fail proof of insulation strength, used on new and repaired equipment.
A dielectric withstand test, also known as a high-potential, hipot or overpotential test, applies a test voltage higher than the equipment will see in service between its energized parts and ground, or between separate circuits, and holds it for a stated time. If the insulation holds without breakdown, flashover or excessive leakage, it passes. The test voltage may be alternating, direct or very low frequency, depending on the equipment and the standard. It is used on switchgear and switchboards, circuit breakers, busway, power and instrument transformers, regulators, reactors, motors and cables, and on the vacuum interrupters inside medium-voltage breakers, where it is the accepted proof that the vacuum is intact. Unlike diagnostic measurements such as insulation resistance or power factor, which describe how good the insulation is, a withstand test answers a simple question: can the insulation stand the required stress. It is a stress test, so it is performed with care and only after lower-voltage tests show that the equipment is a reasonable candidate.
Insulation can look fine in a low-voltage measurement and still contain a defect that breaks down under stress, such as a void, a crack, a contaminated surface, a misplaced barrier or a reduced clearance caused by assembly or damage in transport. Operating voltage is only part of what equipment must survive, since switching operations and lightning produce transient overvoltages. A withstand test proves that there is margin above normal conditions, so a weakness appears on the test bench or in a planned outage and not as an in-service fault. It is therefore used on new equipment after installation, on equipment after repair or reassembly, and in some maintenance programs on equipment whose condition justifies it. For vacuum interrupters it is the standard way to confirm vacuum integrity in the field. The test is deliberately demanding, so it is reserved for cases where proof of strength is worth the risk, and it is normally preceded by diagnostic tests that screen for obvious problems.
Insulation fails when the electric stress across it exceeds what the material and the geometry can withstand. Voids and defects concentrate the stress, so a weak spot breaks down at a lower applied voltage than sound insulation does. A withstand test applies a stress high enough to reveal these weak spots but chosen to stay below the level that would harm good insulation, with the voltage held for a timed interval to allow time-dependent mechanisms to develop. Leakage current is monitored throughout. A small steady current is normal, and a sudden rise, a flashover or a trip is a failure. The three common voltage types act differently. Alternating voltage stresses the insulation as in service and charges the capacitance heavily, so large equipment needs a large source. Direct voltage needs little power but distributes stress by resistance, and it can leave trapped charge in some insulation types. Very low frequency combines modest power with a stress pattern closer to alternating current, which is why it is preferred for extruded cable.
The equipment is de-energized, isolated from every source, grounded and locked out. Because the test applies high voltage, safety is the dominant concern: a controlled test area with barriers or an attendant, clear communication, a defined grounding procedure and a means to discharge the equipment afterward. Everything that must not see the test voltage is disconnected or grounded, including surge arresters, electronic devices, instrument transformers and meters, control wiring and neighboring circuits. Lower-voltage tests, usually insulation resistance, are completed first so that a visibly weak unit is not stressed further. The test set is connected between the part under test and ground, with the other parts grounded, and the voltage is raised smoothly at a controlled rate to the level specified by the governing standard, held for the stated time and brought back down in the same controlled way. Leakage current is watched for the whole interval. The equipment is then discharged through the set and grounded with a hook, and it stays grounded for a period that depends on its capacitance. Test results are accepted only if the equipment passes without breakdown, and a repeat insulation resistance reading afterward confirms it was not harmed.
The record states the equipment identification and the part tested, for example phase to ground or across an open contact gap, the type and level of the test voltage, how long it was held, how the voltage was raised and lowered, the leakage current observed, and the identity and calibration status of the test set. It notes any equipment disconnected or grounded to protect it, the ambient conditions, the result as pass or fail, and any event during the test such as a trip or a corona sound. The insulation resistance readings taken before and after the test are referenced. These details matter because a pass applies only to the configuration and level at which it was achieved, and because a later reviewer needs to know whether the test stressed the insulation in the way the standard intends.
A dielectric withstand test is judged on a simple basis, and this page gives no voltages or durations. The principal criterion is that the equipment holds the specified test voltage for the specified time without breakdown, flashover, sustained discharge or an excessive rise in leakage current. The test level is not a free choice. It is set by the governing standard as a defined fraction of the equipment's factory test level, and it differs for acceptance, where the field level is typically lower than factory, and for maintenance, where it is lower again. It also depends on the type of voltage used and on the type of equipment. A second criterion is the leakage current trend during the hold: a steady value is acceptable, while a creeping or accelerating rise is a warning even if there is no breakdown. A third is the insulation resistance before and after. ANSI/NETA ATS-2025 and ANSI/NETA MTS-2023 give the test requirements by equipment class, and IEEE C57.12.91 addresses dry-type transformers. Refer to your own licensed copies for the levels and durations.
At acceptance the test proves new or repaired equipment against a field level based on its factory level, usually after installation and before energization. In maintenance the level is lower, the test is optional for many equipment types, and the choice depends on the condition of the equipment and the owner's risk tolerance. The equipment class changes the details. Medium-voltage switchgear and breakers are tested phase to ground and across open contacts. Vacuum interrupters are tested across the open gap, with attention to the x-ray emission that a high-voltage vacuum gap can produce. Cables use a voltage type suited to the insulation, with very low frequency preferred for extruded cable and direct voltage historically used on paper-insulated cable. Transformers are normally tested with other methods, or with an applied voltage test only where the standard calls for it. Motors and generators are often tested with direct voltage in a controlled step pattern. In all cases the governing standard for the class defines the procedure.
The equipment classes and standard sections that call for this test, and the UltraDb forms that carry it.
Insulation resistance is a diagnostic reading at a modest voltage that describes how good the insulation is. A withstand test applies a much higher voltage and gives a pass or fail answer on whether the insulation can hold it. The resistance test is normally done first, to screen out weak equipment.
It is a stress test, so there is always some risk, especially with aged or marginal insulation. That is why levels are set below factory test levels for field work, why gentle voltage ramps are used, and why diagnostic tests come first. Equipment that fails would likely have failed in service.
A long cable is a large capacitor, so power-frequency testing needs a very large source. A very low frequency set charges the cable with far less power and stresses extruded insulation in a manner closer to service than direct voltage, which can leave harmful trapped charge in aged polymer cable.
In the tables of the governing standards, for example ANSI/NETA ATS-2025 for acceptance and ANSI/NETA MTS-2023 for maintenance, together with the manufacturer's instructions. Never choose a level by rule of thumb; use your licensed copy of the standard.
UltraDb references ASTM, ANSI, IEEE, NETA, NFPA, NEMA, FAA, TIA, and IEC standards descriptively, to identify the published methods its forms are built from. UltraDb and Verlecta are not affiliated with, endorsed by, certified by, or licensed by NETA, ASTM, ANSI, IEEE, NFPA, NEMA, FAA, TIA, IEC, or Megger. Standard names and section numbers are the property of their respective organizations. Always test to the edition your contract specifies. Where a standard is sold through the ANSI webstore, Verlecta may earn a commission on a purchase made through our link, at no extra cost to you. A commission never decides which standards we cite or how a form is built.