The power-frequency withstand test is a formal dielectric proof test, not an insulation-resistance measurement at a higher setting. It applies a specified AC voltage, waveform/frequency and duration to a precisely configured MV main circuit and judges whether the insulation withstands without disruptive discharge under the applicable IEC procedure.
This practical guide explains factory routine withstand testing of metal-enclosed MV switchgear under IEC 62271-1 and IEC 62271-200. It covers rated withstand voltage, test boundaries, VTs/arresters/sensors, source capacity, voltage measurement, phase-to-earth and phase-to-phase configurations, safety, atmospheric conditions, failure analysis, restoration and the difference between factory and on-site testing.
Executive conclusions
- Take test voltage, duration, configuration and acceptance from the exact contracted IEC/product-standard edition; never improvise from nominal voltage alone.
- The rated short-duration power-frequency withstand voltage is an insulation rating, not the normal operating voltage.
- Define the complete test object and isolate components that are not intended to receive the test—especially VTs, surge arresters, voltage-detection systems, LPITs, cables and electronics.
- Measure the applied high voltage with an approved measuring system; a Variac primary indication or transformer ratio alone is not sufficient evidence.
- Size the source for the test-object capacitance and required voltage without excessive waveform distortion or voltage collapse.
- Bond non-tested phases and the enclosure exactly as the procedure requires; phase-to-earth and phase-to-phase stresses must both be represented where applicable.
- Use a controlled ramp, hold and reduction sequence. Timing starts only when the specified test voltage is established.
- Safety requires an earthed enclosure, controlled test zone, interlocks/communications, discharge and proof of zero residual voltage.
- Preserve the first failure, investigate test setup versus product, and repeat only after approved disposition and regression planning.
- Factory routine, site commissioning and cable tests are different. Do not repeat or combine them without manufacturer and contractual guidance.
1. Purpose and dielectric meaning
The test verifies that the manufactured main-circuit insulation can withstand the specified short-duration alternating-voltage stress. It can expose gross insulation defects, incorrect clearances, contamination, damaged supports, assembly errors, sharp conductive protrusions and unintended connections.
| Evidence | Question answered |
|---|---|
| Rated insulation level/type evidence | Is the design qualified for declared dielectric duties? |
| Routine power-frequency withstand | Was this manufactured assembly free from specified dielectric defects? |
| Insulation resistance | What diagnostic DC leakage/resistance is observed in a defined condition? |
| Partial-discharge test | What discharge activity is present under a defined PD method, where applicable/agreed? |
These tests are complementary. A high IR value or quiet acoustic scan cannot replace the formal routine withstand requirement.
2. Rated voltage, Um and Ud
The applicable IEC tables coordinate the highest voltage for equipment and rated insulation levels, including the rated short-duration power-frequency withstand voltage (commonly denoted Ud) and impulse withstand level. The test value is not calculated simply as a fixed multiple of system voltage.
- Confirm highest voltage for equipment, not only feeder nominal voltage.
- Confirm whether phase-to-earth and across-isolating-distance requirements differ.
- Use the contracted standard edition/amendment and selected insulation level.
- Apply altitude/service-condition provisions exactly as required/agreed.
- Do not substitute lightning-impulse peak values into an AC RMS test.
- Record whether voltage is RMS and how it is measured.
This article intentionally does not reproduce copyrighted IEC voltage tables. The approved test sheet should cite the exact table/clause and enter the project value before testing begins.
3. Define the test object
- complete busbar and feeder main circuit;
- one phase versus bonded other phases/enclosure;
- phase-to-phase configuration where required;
- bus side and cable side with breaker/device closed;
- across open switching-device/isolation distance where applicable;
- separate shipping unit or complete connected lineup;
- circuit breaker/switch component tested under its relevant product procedure;
- solid, gas/fluid or combined insulation state and rated filling pressure/density.
Create a single-line test diagram for each configuration showing HV connection, earth/bonding points, open/closed devices, isolated components and barriers. Verbal descriptions such as “test panel at 38 kV” are inadequate.
4. Components to isolate or treat separately
| Component | Why it matters | Required control |
|---|---|---|
| Inductive VT | Winding/ferroresonance/secondary paths; may not be included in assembly test | Disconnect/withdraw/fuse/isolate per manufacturer procedure |
| Surge arrester | Designed to conduct/clamp overvoltage; test can damage or be limited | Disconnect and restore under controlled checklist |
| VPIS/VDS/coupling capacitor | Capacitive/electronic interface with defined test treatment | Follow device and switchgear instructions |
| LPIT/sensor/merging interface | Electronic/low-power insulation limits | Use approved isolation/test mode |
| Power cable/termination | Adds large capacitance and has different site test standard | Normally define separately; do not include inadvertently |
| RC snubber/capacitor | Increases current and may be overstressed | Review circuit and component rating |
| CT secondary/electronics | Primary test can induce/transfer voltage | Apply specified secondary earthing/shorting and isolation |
Every disconnection creates a restoration hazard. Use a numbered isolation register with before/after photographs and independent check.
5. Test-source capacity
The switchgear appears largely capacitive at power frequency. Approximate charging current is:
Ic = 2πfCV
and apparent power is approximately:
S ≈ V × Ic = 2πfCV²
- Estimate/measure total capacitance for the exact connected lineup.
- Include test leads, coupling components and any connected cable.
- Ensure source kVA and regulation maintain the specified voltage during the hold.
- Verify frequency and waveform requirements under IEC 60060-1/product standard.
- Avoid severe distortion, resonance or voltage magnification.
- Use current limiting/protection that safely clears a failure without invalidating the applied waveform.
- Confirm discharge arrangement is rated for stored energy.
Because required kVA rises with the square of voltage, doubling test voltage roughly quadruples capacitive kVA for unchanged capacitance/frequency.
6. High-voltage measurement system
- Use an approved divider/transformer/measuring system suitable for AC RMS and the voltage range.
- Record make/model/serial, scale factor, calibration and uncertainty.
- Measure at the high-voltage test point or through a validated measuring chain.
- Verify lead/layout effects, clearances and earthing of the divider.
- Check the complete system before test with an approved functional/ratio verification.
- Record actual applied voltage during the hold, not only the setpoint.
- Define the conformity decision rule if measurement uncertainty is material near tolerance.
IEC 60060-2:2025 is the current high-voltage measuring-system standard. Product standards can specify additional requirements; the laboratory’s approved system must satisfy the applicable combination.
7. Test-cell safety
- Approved risk assessment, method statement and competent authorised test personnel.
- Physical fenced exclusion zone with controlled access and warning indicators.
- Visible/tested emergency stop and source interlock.
- Reliable communication between test controller, observer and equipment operator.
- All accessible enclosure parts earthed; defined single test return path.
- Doors/covers in the test configuration and mechanically secured.
- Remote operation where practical; no person inside the danger zone during energisation.
- Discharge stick/earthing device, voltage detector and rescue provisions appropriate to the facility.
- Pre-test call/clearance and post-test earth/prove-dead sequence.
- Control of induced voltage in adjacent circuits and test leads.
Never depend on software alone to protect personnel. The physical test-cell interlock and earthing system must remain effective if the control computer freezes.
8. Pre-test readiness
- Construction, clearances, partitions and foreign-material inspection complete.
- Insulation clean, dry and undamaged; gas/fluid condition within specified limits.
- Bus joints/supports and enclosure earth complete.
- Correct breaker/disconnector/earthing-switch positions confirmed.
- VTs, arresters, cables, sensors and electronics isolated per approved register.
- CT secondary circuits in specified safe state.
- Test source, divider, protection and discharge system checked.
- Atmospheric/environmental conditions recorded and correction assessed if applicable.
- Optional diagnostic IR pre-check reviewed; anomalous result resolved.
- Approved test sheet already contains voltage, duration, phases and acceptance.
9. Phase-to-earth test configuration
- Connect HV to the phase or bonded phase group defined by the standard/procedure.
- Bond non-tested phases, main earth bar and enclosure to test return/earth as required.
- Close switching devices needed to include the complete path; open/isolate parts for separate tests where specified.
- Maintain adequate external test-lead clearance so a laboratory lead flashover does not masquerade as product failure.
- Repeat configurations for all phases/sections required by the standard.
- Document any phase grouping permitted by the procedure and why it covers the required stresses.
10. Phase-to-phase and across-isolating-distance stress
Depending on the insulation arrangement and product requirements, testing must represent phase-to-phase insulation and/or the open isolating distance. Do not assume a phase-to-earth test automatically stresses every interface correctly.
- Identify which terminals are energised and which are bonded to return.
- Use the required switching-device open/closed position and mechanical interlocks.
- Ensure an open gap does not leave a floating conductor at uncontrolled potential.
- Check capacitive coupling and divider/reference arrangement.
- Use the specific voltage required across isolating distance—do not infer it from the normal phase-to-earth value.
- Record each test state with a diagram/photo.
11. Voltage application sequence
- Confirm test-cell clear, access locked, earths/configuration checked and personnel ready.
- Energise the source at low/zero output.
- Raise voltage smoothly according to the approved rate/procedure while observing voltage/current.
- Stop and discharge on abnormal noise, current, corona location, smoke or instability.
- Establish the specified test voltage within permitted tolerance.
- Start the required duration only after voltage is established.
- Monitor/record voltage, current, time and observations through the hold.
- Reduce voltage smoothly to the specified low/zero level before opening the source.
- De-energise, discharge, apply earth and prove residual voltage absent.
- Announce test complete before access; inspect the tested section.
- Change configuration only under the written sequence and repeat.
Do not count ramp time as the specified hold unless the governing procedure explicitly permits it. Do not keep equipment at full voltage longer merely to allow witnesses to take photographs.
12. Acceptance and observations
- No disruptive discharge/breakdown on the test object under the specified test.
- Specified voltage maintained for the required duration within tolerances.
- No unacceptable protective operation or voltage collapse attributable to the product.
- Any discharge external to the test object identified and test validity assessed.
- Post-test inspection shows no damage, tracking, puncture, carbonisation or displaced parts.
- Isolated components restored and functional checks completed.
- Exact product-standard acceptance wording governs; observations such as audible corona are interpreted within that framework, not by guess.
A test-lead flashover can invalidate the test without proving the switchgear failed. Conversely, calling a flashover “only external” requires objective evidence of its location and a valid repeat after correcting the setup.
13. Failure investigation
- Trip/de-energise, discharge, earth and secure the area.
- Preserve waveform/voltage/current/time and first-failure observations.
- Identify whether the event was test source, external lead, divider, environment or product.
- Inspect for arc marks, tracking, loose hardware, contamination, damaged insulation and unintended connected devices.
- Segment phases/compartments only under an approved diagnostic plan.
- Open an NCR; quarantine potentially common affected units.
- Obtain design-authority disposition and repair procedure.
- Assess whether repair changes type-test applicability or adjacent insulation.
- Repeat the failed configuration and required regression tests; do not erase failure history.
- Update drawings, isolation/restoration register and final report.
Repeatedly reapplying full voltage to “burn away” a defect is unacceptable. Each failure can carbonise a path, damage sound insulation or obscure the root cause.
14. Atmospheric conditions and altitude
Air-insulation withstand is affected by atmospheric conditions. IEC 60060-1 provides high-voltage test-technique treatment and IEC 62271/common specifications address service conditions. Apply atmospheric correction only where the applicable product test permits/requires it and record pressure, temperature and humidity.
- Do not reduce factory voltage ad hoc because the equipment will operate at altitude.
- Confirm design altitude correction/derating and insulation coordination separately.
- Record whether insulation is internal sealed/gas/solid versus external air.
- Ensure test-room clearances and test leads are adequate at the actual conditions.
- Use the contracted standard/manufacturer engineering decision for non-standard service conditions.
15. Factory versus on-site testing
| Factory routine test | On-site/commissioning test |
|---|---|
| Controlled assembly before shipment | Installed lineup, cables and site environment |
| Product-standard routine value/method | Agreed site method/value; may differ |
| Test cell with qualified source/measurement | Portable source, greater noise/clearance constraints |
| Cables/field equipment normally excluded unless specified | Need explicit separation of switchgear and cable tests |
| Confirms manufacture | Finds transport, assembly, contamination and termination issues |
Do not subject VTs, arresters or switchgear to a cable test waveform/voltage by accident. Establish isolating gaps, test adapters, earthing and manufacturer limits in the commissioning method.
16. Test report
- project, lineup/panel/serial and rated insulation data;
- standard edition, clause/table and approved procedure revision;
- test diagram for every configuration;
- device positions and isolated/connected component register;
- test voltage, frequency, waveform parameters as required and duration;
- source/divider/measuring-system IDs, calibration and uncertainty;
- ambient pressure/temperature/humidity and applied correction;
- voltage/current/time record and observations;
- pass/fail criterion and authorised result;
- failure/NCR/repair/retest history;
- discharge and final link/device restoration verification.
Common mistakes
- Calculating test voltage from nominal voltage instead of the IEC insulation level.
- Leaving a VT, arrester, VPIS or cable connected unintentionally.
- Reading voltage from the source primary only.
- Using a source too small for capacitive kVA.
- Incorrectly floating non-tested phases.
- Counting ramp time as the full hold.
- Allowing external test-lead flashover and calling the product passed.
- Reapplying voltage repeatedly without failure analysis.
- Skipping discharge/return-voltage control.
- Combining a cable and switchgear site test without interface review.
Official standards and primary references
- IEC 62271-1:2017+AMD1:2021 — current common rated-insulation and dielectric test requirements.
- IEC 62271-200:2021+AMD1:2024 — current routine dielectric requirements for metal-enclosed MV assemblies.
- IEC 62271-100:2021+AMD1:2024 — current circuit-breaker dielectric requirements.
- IEC 60060-1:2025 — current general terminology and high-voltage test requirements.
- IEC 60060-2:2025 — current requirements for high-voltage measuring systems.
- ISO/IEC 17025:2017 — testing/calibration competence and traceability framework where applicable.
Engineering note: The exact IEC tables, tolerances, duration and circuit configurations must be read from licensed purchased standards. This article provides the execution framework and deliberately does not invent substitute voltage values.