High Voltage Tester

What is a High Voltage Tester

A high voltage tester, also called a hipot tester (a contraction of "high potential") or dielectric withstand tester, is an instrument that deliberately applies a voltage well above a device's normal operating voltage across its insulation system and observes whether the insulation holds. The test verifies that live conductors stay safely isolated from accessible metal surfaces and from each other under abnormal stress, so that a fault, a surge, or simple ageing does not turn an enclosure into a shock or fire hazard. Unlike a digital multimeter or a clamp meter, a high voltage tester is a stress-and-judge instrument: it produces a controlled overvoltage and returns a pass or fail verdict based on breakdown and leakage current.

Functionally the tester has three parts: a high-voltage source (a step-up transformer for AC, or a transformer plus rectifier and smoothing for DC) that generates the test potential; a current-sensing and protection circuit that measures the leakage flowing through the device under test and trips the output the instant it exceeds a set limit or an arc is detected; and a controller that ramps the voltage up at a defined rate, dwells for the test time, ramps down, and discharges any stored energy. The discharge step matters: a tested cable or capacitor can retain a lethal charge after a DC test, so safe testers bleed the device down through an internal resistor before the operator can touch it.

It is important to distinguish the high voltage tester from its close relatives. An insulation resistance tester, or megohmmeter, applies a lower steady DC voltage, typically 250 V to 5000 V, and reports a resistance in megohms or gigohms, a quantitative number describing how good the insulation is at that moment. A high voltage tester instead asks a binary question: will the insulation survive this stress voltage. Many bench instruments fold both functions into one unit, measuring insulation resistance first and then ramping into the withstand step, but the two measurements answer different engineering questions and should not be confused.

The need for high voltage testing is old. As mains electricity spread through factories and homes in the early twentieth century, withstand testing of transformers, cables and machines became standard factory practice, and the principle was codified internationally as the IEC 60060 series of high-voltage test techniques, whose first parts date to the mid twentieth century and whose current third edition was published in 2010 with a fourth edition issued in 2025. In parallel, the rise of mass-produced consumer and IT equipment after the 1960s created the production-line hipot tester, a fast benchtop instrument designed to test thousands of units a day against safety standards such as the IEC 60950 family, now superseded by IEC 62368-1.

The scale of voltages involved spans more than two orders of magnitude. A benchtop tester checking a power supply works at 1.5 to 3 kV; a field DC hipot proving a 35 kV cable runs at 50 to 80 kV; and an impulse generator validating an extra-high-voltage transformer can reach well over 1000 kV peak. Each band uses different source topologies, different insulation around the tester's own high-voltage parts, and different operator-safety regimes, which is why no single instrument covers the whole field and why selection always begins by fixing the voltage class of the device under test.

Instrument Types and Classification

High voltage testers split first by application into two broad worlds. The first is product-safety testing, where compact bench or rack instruments screen finished goods (power supplies, appliances, motors, medical devices) at 1 to 5 kV AC or 6 to 15 kV DC, optimized for speed, automation and operator throughput. The second is power-apparatus field and factory testing, where large mobile or fixed test sets prove the dielectric strength of cables, transformers, switchgear such as a circuit breaker, and insulators at 10 kV to several hundred kV. The table below maps the main instrument types across both worlds.

Bench hipot testers are the workhorses of the production line. An AC-only model such as the SCI 295 outputs up to 5 kV AC at 12 mA; a DC-only model such as the SCI 294 reaches 6 kV DC at 5 mA; and a high-power AC model such as the SCI 298 supplies 5 kV at 100 mA from a 500 VA source for high-capacitance loads. HIOKI's automatic withstanding HiTESTER 3174 generates up to 5 kV AC by a PWM method that holds output independent of the mains voltage, while the combined 3153 adds insulation resistance up to 9999 megohms. These instruments are built for repeat testing with pass or fail lamps, scanners and PLC handshakes.

DC hipot and insulation testers handle loads that an AC source cannot. A long power cable or a large AC motor winding presents a large capacitance; an AC test would draw a continuous charging current that swamps the leakage reading or trips the tester. DC charges that capacitance once and then measures only true leakage, so DC is the practical choice for cables, capacitor banks and large windings. Field DC sets such as the High Voltage Inc PTS and DBT series span 37.5 kV to 600 kV DC at up to 10 mA, with the higher voltages built as two-piece controller-plus-tank assemblies.

AC test sets and resonant systems apply 50 or 60 Hz stress, which reproduces real service conditions and exercises insulation in both polarities. A plain power-frequency set is a step-up transformer that must supply the full reactive charging current of the load, so for large objects it becomes impractically heavy. A resonant system instead tunes a high-voltage reactor against the test-object capacitance, so the supply feeds only the small real losses; this is the standard approach for cable, GIS and transformer testing. VLF sets shift the AC frequency down to 0.1 Hz to shrink the charging current, making them the dominant field tool for medium-voltage polymer cables, and they double as the source for tan delta and partial discharge diagnostics.

Test Methods: AC, DC, Resonant, VLF

The choice of test waveform is the single most consequential decision in high voltage testing, because each method stresses insulation differently and interacts differently with the capacitance of the device under test. The four mainstream methods are AC at power frequency, DC, AC at resonance, and AC at very low frequency. The table compares them on the engineering axes that matter for selection.

AC at power frequency is the reference method because it reproduces the actual service stress on insulation: the voltage reverses polarity 100 or 120 times a second, exercising the dielectric in both directions and revealing defects that a one-directional DC stress can miss. Its drawback is the charging current. Capacitive current rises with frequency and with load capacitance, so testing a large cable or a GIS bus at 50 Hz demands a transformer that can source tens of amps of reactive current, which is heavy, costly and power-hungry. This is why a plain AC set is reserved for small objects, oil cups, bushings and laboratory type tests.

DC testing sidesteps the charging-current problem. After an initial inrush that charges the load capacitance, the steady-state current is pure leakage, so a small source can hold high voltage on a very capacitive object. DC is therefore standard for proving cables and capacitor banks and for the steady-state insulation resistance measurement. Two cautions apply. First, DC and AC stress insulation differently, and the conventional DC test voltage is set to 1.414 times the equivalent AC value so the DC peak matches the AC peak. Second, DC injects space charge into modern extruded polymer cable insulation, which can accelerate ageing and trigger premature in-service failures, so DC withstand testing of XLPE and EPR cables is now discouraged in favour of VLF.

AC resonant systems resolve the charging-current dilemma without leaving the AC domain. A high-voltage reactor is tuned so its inductance resonates with the test-object capacitance at or near power frequency; at resonance the reactive currents of the reactor and the load cancel, and the supply only has to feed the real dielectric and copper losses, typically a few percent of the reactive power. A resonant set therefore delivers the same kV from a fraction of the input power, weight and footprint of a transformer set, which is why series and parallel resonant systems are the standard tool for routine and type testing of cables, GIS and large power transformers.

VLF testing attacks the same problem from the frequency axis. Because charging current is proportional to frequency, dropping from 50 Hz to 0.1 Hz cuts the charging current by roughly five hundred times, so a compact, portable 0.1 Hz set can withstand-test a long medium-voltage cable that would need a truck-sized 50 Hz transformer. VLF still applies a bipolar AC stress, so unlike DC it does not deposit harmful space charge in polymer cables. The two common output waveforms are sinusoidal and cosine-rectangular; only a true sine-wave VLF source can also perform tan delta and partial discharge diagnostics, where the dissipation factor at 0.1 Hz gives an especially sensitive read on insulation ageing.

Standards and Test Voltage Formulas

High voltage testing is one of the most heavily standardized areas in electrical engineering, because the test voltage, the duration and the pass criterion are not free choices: they are dictated by the safety or apparatus standard that applies to the device under test. The instrument generates the voltage, but the standard sets the number. Two layers of standards apply: the high-voltage test technique standards that define how a valid test is run, and the product or apparatus standards that define what voltage to apply.

IEC 60060-1 is the master high-voltage test technique standard. It applies to dielectric tests on equipment whose highest voltage exceeds 1 kV AC (1.5 kV DC in the 2025 edition) and defines the AC, DC, impulse and combined test procedures, the permitted measuring uncertainties, and the tolerances on the test voltage. The companion IEC 60060-2 covers the measuring systems, and IEC 62475 now carries the high-current measurement methods that earlier editions kept inside IEC 60060-1. Any laboratory claiming a traceable HV test result references this family.

For product safety, the governing standard depends on the product class. IEC 62368-1, which has replaced the long-serving IEC 60950-1 for IT equipment and IEC 60065 for audio and video, sets the dielectric withstand voltage for hazard-based safety. IEC 61010-1 covers electrical test, measurement and laboratory equipment, and its third edition notably prohibits DC withstand testing on circuits that are normally energized by AC. IEC 60601-1 covers medical electrical equipment, where patient-protection insulation is tested to higher voltages than ordinary mains insulation. The table summarizes the dielectric withstand logic of the most common product-safety regimes.

The most widely used production formula is 2 times the working voltage plus 1000 V AC for basic and supplementary insulation. For a 230 to 240 V mains product this yields roughly 1500 V AC; reinforced insulation, which combines basic and supplementary into a single barrier, is tested higher, near 3000 V AC. Circuits at or below 50 V working voltage use a fixed 500 V AC, while circuits above 300 V use 2 times working voltage plus 1840 V AC, capped near 3000 V AC. When a standard permits substituting DC, the DC test voltage is set to 1.414 (the square root of 2) times the AC value so the DC level matches the peak of the AC waveform.

Field cable testing follows a separate guide family. IEEE 400 and its part document IEEE 400.2 define VLF withstand levels expressed as multiples of U0, the rated rms phase-to-ground voltage, ranging from below 2 U0 up to 3 U0 for cables rated 5 to 69 kV, with durations of about 30 minutes for maintenance tests on aged cable and up to 60 minutes for acceptance tests on new cable. Insulation resistance and polarization index follow IEEE 43, where the polarization index is the ratio of the 10-minute to the 1-minute resistance reading, a PI above 2 indicating sound winding insulation and IEEE 43 advising that readings above 5000 megohms (corrected to 40 degrees C) need not be index-tested.

Key Specification Parameters

Reading a high voltage tester datasheet means looking past the headline kV figure. The same nominal output can hide very different real capability depending on the current rating, the trip logic and the metering resolution. Eight parameters drive selection: maximum test voltage, output current and power, trip current and arc detection, ramp and dwell timing, voltage and current accuracy, insulation resistance range, output regulation, and operator-safety features. Each is explained below.

Maximum test voltage is the headline number, but it must be read together with the current rating because a high-capacitance load can pull the voltage down if the source runs out of current. A 5 kV AC tester rated at 12 mA cannot hold 5 kV into a load that demands 40 mA of charging current; the output simply collapses or trips. For high-capacitance work, choose a source whose volt-ampere rating covers the worst-case charging current, for example a 500 VA AC source delivering 100 mA at 5 kV.

Output current and power set the size of the load the tester can drive. Bench DC testers commonly supply 5 to 20 mA, enough for dry, low-capacitance assemblies; field DC hipots reach 10 mA at hundreds of kV, sufficient because steady leakage is small once the load is charged. AC sources are rated in VA because the load is mostly reactive: a 500 VA set is a typical dividing line between general-purpose and high-capacitance bench testers.

Trip current and arc detection define the pass or fail criterion. The trip current is the leakage threshold above which the tester declares a fail; it is set just above the legitimate capacitive plus resistive leakage of a known-good unit, with margin for production spread. Product standards frequently cite resistive-leakage limits near 0.25 mA AC and 0.1 mA DC. A separate arc-detection or flashover limit catches a momentary breakdown even when the average current stays below the trip point, which is essential for finding marginal insulation that flashes over only briefly.

Ramp and dwell timing govern how the stress is applied. A controlled ramp, often limited to about 500 V per second, avoids the inrush current spike that a sudden voltage step would draw into the load capacitance and that could nuisance-trip a good unit. The dwell time is the standard-mandated hold at full voltage, typically 60 seconds for routine testing or as short as 1 second at 120 percent voltage for fast production screening, followed by a controlled ramp-down and a discharge of stored energy.

Voltage and current accuracy determine whether a borderline unit is judged correctly. Quality bench testers specify output-voltage accuracy on the order of plus or minus 1 to 2 percent of setting and leakage-current resolution down to microamps or better, with the HIOKI ST5680A DC tester, for example, resolving leakage to 0.001 microamp. Poor metering resolution blurs the line between a marginal pass and a true fail. Insulation resistance range on combined instruments commonly spans up to several thousand megohms; the HIOKI 3153, for instance, reads insulation resistance to 9999 megohms alongside its 5 kV withstand function.

Output regulation describes how steadily the tester holds the set voltage as the load or the mains supply varies. Instruments that derive their high voltage by a regulated PWM method, such as the HIOKI 3174, hold output independent of mains fluctuation, which matters on factory floors with noisy supplies. Finally, operator-safety features (the tester is itself a hazardous-voltage source) include dual-button start, light-curtain or fixture interlocks, automatic discharge of the device under test, and compliance of the instrument and its probes with IEC 61010-1 and IEC 61010-2-031. These are not optional extras: the tester is a deliberate shock hazard and its own safety interlocks protect the operator.

Selection Decision Factors

To turn the preceding five chapters into a specific model, follow the decision sequence below. The order matters: most selection mistakes come from fixing on a brand or a kV figure before the governing standard and the load type have been pinned down. These eight steps work as a fixed RFQ template for both bench and field instruments.

1. Identify the governing standard first: the device under test, not the tester, decides the test voltage, duration and pass criterion. Fix IEC 62368-1, IEC 61010-1, IEC 60601-1, IEC 60335-1, IEEE 400.2 or the relevant apparatus standard before anything else, because it sets the minimum kV, the AC-versus-DC requirement and the trip limit.
2. Set the required test voltage with margin: derive the test kV from the standard formula (for product safety, 2 x working voltage + 1000 V AC for basic insulation, higher for reinforced), then choose a tester whose maximum output comfortably exceeds it. Do not buy a tester whose rated voltage is the exact test voltage with no headroom.
3. Choose AC, DC or VLF by load type: AC for general product safety and real-condition stress; DC for high-capacitance loads, batteries and capacitor banks; VLF for medium-voltage polymer cables; AC resonant for large cables, GIS and transformers. Remember IEC 61010-1 forbids DC on circuits normally energized by AC.
4. Size the current and VA rating to the load capacitance: compute the worst-case charging current of the device under test and confirm the source can supply it at full voltage. A high-capacitance load needs a 100 mA, 500 VA class AC source; a dry low-capacitance assembly is fine on a 5 to 12 mA source.
5. Set trip current and arc detection: specify a leakage trip just above good-unit leakage with production margin, and require an independent arc or flashover detector so brief breakdowns are caught. For combined analyzers, decide whether you also need ground-bond and insulation-resistance steps in the same cycle, or whether a separate earth ground tester better suits installed-equipment checks.
6. Specify metering, ramp and timing: require the voltage and current accuracy, leakage resolution, programmable ramp rate (around 500 V per second) and dwell time the standard demands, plus controlled ramp-down and automatic discharge of the device under test.
7. Match the operator-safety and automation interface: dual-button start, fixture or light-curtain interlock, discharge indication, and IEC 61010-1 / IEC 61010-2-031 compliance for the instrument and probes; for production, add scanners, PLC or LAN handshakes, barcode logging and pass-fail signalling.
8. Total cost of ownership: purchase price plus annual calibration (a high voltage tester must be calibrated against a traceable reference, often yearly), fixturing, operator training, and for field sets the transport, weight and setup time. A resonant or VLF set costs more upfront than a transformer set but is lighter and cheaper to move and power over its life.

One last commonly overlooked dimension is manufacturer serviceability and calibration support: local calibration laboratories, traceable certificates, spare high-voltage modules, and firmware or fixture support. Bench instruments from HIOKI, Chroma, SCI and Associated Research (Ikonix), Vitrek and Sefelec are built for high-throughput production lines with scanners and automation; field sets from High Voltage Inc, Megger, OMICRON, BAUR, b2 electronics and HV Technologies prioritize kV reach, portability and diagnostic add-ons such as tan delta and partial discharge. Confirm that whichever you choose has a calibration path and spare-parts support in your region before committing, because a high voltage tester out of calibration is a compliance liability, not just an inconvenience.