Selecting a high voltage withstand test method means matching the test objective, the asset and its capacitance, the governing standard and the site constraints to one of the available method families: power-frequency AC, resonant AC, very low frequency (VLF), DC and impulse.
No single method is universally best, because each applies a different stress, suits different assets and proves a different aspect of insulation integrity. The selection framework starts with the question the test must answer, then narrows the method family, then sizes the equipment for the actual load and site.
Deciding the Withstand Test Objective First
The first decision is not which equipment to use but what the test must prove. A withstand test answers one question: can this insulation system hold the specified voltage for the specified time without breakdown? That question appears in different forms across the asset life: proving a new cable circuit after installation, verifying a transformer after transport, qualifying a design in the factory, or checking aged insulation before returning an asset to service. The objective changes the method family, the voltage level, the duration and the acceptance logic, and it should be written down before any equipment is discussed.
Distinguish withstand testing from diagnostic testing. A withstand test is a proof: it passes or fails on breakdown, and a pass does not grade the insulation or reveal incipient deterioration. Diagnostic tests such as insulation resistance, tan delta, partial discharge measurement or frequency response analysis measure condition and can warn about problems before they cause breakdown. Withstand testing and diagnostics are often combined, but they answer different questions, and mixing them in one decision produces a programme that neither proves nor diagnoses well. The framework in this article selects the withstand method; the diagnostic layer is added where the objective requires it.
Method Families: Power-Frequency AC, Resonant, VLF, DC and Impulse
Power-frequency AC withstand testing applies a voltage at the operating frequency, usually 50 or 60 Hz, and represents the closest approximation of the steady-state stress the insulation experiences in service. Its limitation is practical: charging the capacitance of a large cable or a large machine at power frequency requires a large reactive power supply, which is why power-frequency testing is often confined to equipment with moderate capacitance or to factory settings where the supply is available.
Resonant AC testing solves the power problem by tuning a reactor to the capacitance of the load. In series resonance, the reactive current circulates between the reactor and the test object, and the supply only provides the active losses, so a comparatively small source can test a highly capacitive load. Resonant systems are used where power-frequency stress is required on large capacitance, such as long cable circuits, GIS, large transformers and generator stators.
Very low frequency (VLF) testing applies an AC voltage at frequencies around 0.1 Hz. The low frequency dramatically reduces the charging current for capacitive loads, allowing a portable source to test cables and other high-capacitance assets in the field. VLF is a withstand method with its own standard context, and it is not interchangeable with power-frequency testing on a one-to-one voltage basis; its use is defined by the governing guidance for the asset and the test objective.
DC withstand testing applies a steady DC voltage and is appropriate for some assets and applications, but it stresses insulation differently from AC. DC does not reproduce the alternating stress distribution across capacitive and resistive layers, and for some insulation systems, particularly aged extruded cables, DC testing carries risks that must be evaluated against the alternative methods. DC remains relevant for certain equipment classes and for specific applications defined by standards or manufacturer instructions.
Impulse testing applies transient voltages that simulate lightning and switching surges. It is used to prove the insulation against steep-fronted overvoltages, typically as a type test on transformers, reactors and other apparatus, and it requires a completely different source: an impulse generator that produces defined waveforms such as the standard lightning impulse. Impulse testing belongs to the factory and design-validation world more than to routine site withstand testing, and its detail is covered in the impulse basics article in this series.
Matching the Method to the Asset and Its Capacitance
Asset type is the strongest single input to the method decision, because it determines the insulation system, the capacitance and the stress that matters. Cables are dominated by their capacitance, which grows with length, and are commonly tested with VLF or resonant sources in the field. GIS and substation equipment are often tested with power-frequency or resonant AC to reproduce the service stress, sometimes with partial discharge measurement added. Transformers have their own test culture: routine dielectric tests follow the transformer standard, and impulse testing validates the design, while site verification after transport uses the checks defined by the commissioning specification. Generator stators combine high capacitance with a requirement for power-frequency-like stress, which makes resonant systems the practical choice.
Capacitance is the engineering quantity behind the equipment decision. The charging current at a given frequency is proportional to the capacitance and the voltage, so the load capacitance determines how much reactive power the source must deliver. Before sizing equipment, estimate the capacitance of the test object: cable capacitance per unit length times the circuit length, or the phase-to-ground capacitance of a machine from its manufacturer data. The estimate feeds the decision between a direct source and a resonant or VLF source, and it is the first number in any equipment specification.
Standards Context and Where Each Method Is Authorised
Withstand testing is governed by standards and guides that define where each method is authorised, what voltage is applied and what constitutes a pass. The IEC 60060 family covers high-voltage test techniques, including general definitions and measurement requirements. Asset-specific standards define the test programmes for transformers, cables, switchgear and machines, and field guides such as the IEEE 400 series for shielded power cables give the context for VLF, DC and diagnostic testing of cable systems. Maintenance specifications such as ANSI/NETA MTS provide the programme framework for maintenance testing in industrial and commercial installations.
The discipline is to verify scope before citing any standard. A standard covers particular assets, voltage classes and test types, and its edition matters because guidance changes between revisions. Before a method is selected, confirm three things: the document that governs the asset and test objective, the clause that authorises the method, and the voltage and duration it defines for your asset class. A method is only “authorised” when the governing document says so for your exact case.
Site Constraints: Power, Size, Safety and Outage
Site constraints often decide between methods that are technically equivalent. The first is power: a field location may not have the supply capacity for a large power-frequency source, which pushes the decision toward a resonant or VLF system that draws less from the site supply. The second is physical size and access: a resonant reactor train or a large impulse generator must reach the test object and be set up within the outage window, while a portable VLF source may be carried to the asset. The third is the outage itself: the test duration, the time to discharge and re-connect, and the availability of the asset all belong in the plan.
Safety constraints are not optional inputs. The method determines the energy stored in the test circuit, the discharge behaviour and the hazard zone, and the safety plan must be built around the actual source and test object rather than copied from another job. High-voltage testing requires controlled access, verified isolation, visible grounding and a discharge procedure matched to the source type. The safety and discharge discipline for withstand testing is covered in the hipot safety article in this series and applies to every method selected through this framework.
When to Combine Withstand Testing with Diagnostics
Withstand testing and diagnostics are complementary when the objective demands both. A monitored withstand test adds partial discharge measurement during the voltage application, so that the test can reveal a defect that would not yet cause breakdown; the combined result says both “held the voltage” and “no significant discharge activity was seen”. Diagnostics before a withstand test can also inform the decision: if insulation resistance or tan delta testing indicates a moisture problem, applying a full withstand voltage may be the wrong next step until the condition is understood.
The combination decision belongs in the test objective. For a new cable circuit, a monitored withstand test may be specified to catch installation defects early. For an aged asset being evaluated for continued service, diagnostics first, with withstand testing only where the condition supports it, is often the safer sequence. Each combination changes the equipment list and the acceptance criteria, so it should be decided before the method and source are selected.
Building a Method-Selection Matrix for Your Specification
| Asset / objective | Power-frequency AC | Resonant AC | VLF | DC | Impulse |
|---|---|---|---|---|---|
| Cables — installation and acceptance | Rarely practical on long circuits | Where power-frequency stress is specified | Common field choice under IEEE 400-series guidance | Application-dependent, with risk notes for extruded insulation | Not used for routine cable acceptance |
| GIS and substation equipment | Factory and site where supply allows | Common for high-capacitance site tests | Less common | Limited | Design/type context |
| Transformers | Factory routine/type dielectric tests per transformer standards | Site AC tests where specified | Not the transformer standard method | Limited, per specific guidance | Type test for lightning impulse withstand |
| Generator stators | Where supply is adequate | Practical choice for large capacitance | Not typical | Application-dependent | Not for stator withstand |
Use the matrix as the starting point of a project specification, then replace every entry with the requirement from the governing standard and the actual asset data. The matrix is a decision aid, not a substitute for the standards that authorise each method.
Frequently Asked Questions
Which withstand method is best for cables?
There is no single best method for all cables. VLF is a common field choice for medium-voltage cable acceptance and maintenance under IEEE 400-series guidance, resonant AC is used where power-frequency stress is required on long circuits, and DC is application-dependent with specific risk notes for extruded insulation. The governing standard and the cable type decide the method.
Why is resonant AC used instead of a direct test transformer?
A resonant source tunes its reactor to the capacitance of the load so that the reactive current circulates in the resonant circuit and the supply provides only the active losses. This allows a comparatively small source to test a highly capacitive load at power frequency, where a direct test transformer would need a much larger supply.
Can a withstand test tell me the condition of the insulation?
A withstand test is a proof, not a diagnostic. It shows whether the insulation held the specified voltage; it does not grade the insulation or reveal incipient deterioration. To assess condition, add diagnostic measurements such as insulation resistance, tan delta or partial discharge, either before or during the withstand test.
Use the framework to build your test specification and explore the high voltage insulation and withstand testing hub on the HVTesters withstand testing page. For transformer-specific test planning, see What Tests Are Required for a Power Transformer.