A VFD-fed motor fails from repetitive voltage peaks across its insulation, not from overload heat, and the deciding quantity is the peak voltage at the motor terminals relative to the corona inception voltage of the stator insulation. Every fix below either changes one of those two numbers or does not.
Fixes that miss the cause
Five explanations and fixes come up repeatedly for VFD-related motor insulation failures. Each one addresses the wrong quantity.
| Common move | Why it fails |
|---|---|
| Buy the motor with the "special wire" and treat the wire as the differentiator | Wire is one input. The inverter duty label is earned when the finished stator passes a corona test at a defined voltage. A stator that fails that test but passes the other leakage tests ships as standard duty. Wire type does not guarantee the stator passed. |
| Declare every non-inverter-duty motor a certain casualty | Many motors not designed for inverter duty run on VFDs for years. Stress depends on bus voltage, cable length and insulation condition, so the outcome varies by installation. |
| Assume a 600 V-rated winding has unlimited margin | Breakdown is set by electric field strength, which depends on geometry, aging and moisture, not by the voltage rating alone. A poorly built winding, or one with shifted insulation sheets, has a lower withstand than the label implies. |
| Blame inductive kickback when the drive stops | The stress is the repeated pulse edges during operation, amplified by cable resonance, not a single turn-off event. A spark to ground carbonizes insulation and eventually leaves a conductive path, but PWM edges drive that discharge many thousands of times per second. |
| Attribute motor failure to mains spikes | Unfiltered 240 V mains can see spikes near 600 V. The VFD rectifier and DC bus stand between the mains and the motor, so those spikes stress the drive's rectifier, not the winding. A severe spike can fail the input rectifier. |
Corona inception voltage as the inverter duty criterion
Corona is a partial discharge that starts just below the arc voltage. The voltage at which it begins is a good predictor of failure from spikes and transients. It also reveals how well the insulation sheets stayed in position during winding, because winding forces often shift insulation and reduce the withstand voltage a little.
Motor manufacturers test finished stators for corona discharge. A stator that passes goes into a motor labeled inverter duty. A stator that fails corona but passes leakage and the other tests becomes a standard duty motor. This sorting explains why two motors built with the same wire can carry different labels.
Older motors with thicker insulation are often resistant to corona and make good inverter duty candidates, although the label will not say so. Judge them on measured insulation condition, not on age.
Bus voltage, pulse amplitude and cable resonance
A PWM drive output switches between the DC bus rails. The bus peak is the rectified line peak: Vbus = VLL x sqrt(2). Pulse amplitude at the drive terminals equals this bus voltage, and cable reflection can raise the amplitude at the motor terminals above it.
| Quantity | Value | Basis / where to read it |
|---|---|---|
| DC bus peak, 230 V supply | 230 x 1.414 = about 325 V | Derived. Read the bus on the drive's DC bus terminals or in the drive's bus-voltage monitor parameter. |
| DC bus peak, 480 V supply | 480 x 1.414 = about 679 V | Derived, assuming a 480 V nominal line-to-line supply. |
| Full-reflection bound at the motor, 230 V system | up to about 650 V (2 x 325 V) | Theoretical upper bound before ringing overshoot. Measure at the motor terminals with a fast scope and a differential probe. |
| Full-reflection bound at the motor, 480 V system | up to about 1,358 V (2 x 679 V) | Same bound, same assumption. Compare against the motor's stated spike withstand and corona inception data. |
| Hand-held meter on the drive output | Reading of 280 V on a 230 V drive | See the measurement note below. |
Longer cables allow resonances that produce peaks considerably higher than line voltage. Cable length is therefore the second decision quantity after bus voltage, and it is the one most often unrecorded on site.
The 230 V case is the benign one. Most standard-wound motors use magnet wire rated 600 V, and a 325 V bus leaves substantial margin even after reflection. A 480 V system has roughly twice the pulse amplitude and stresses insulation much harder, which is why voltage-stress failures concentrate there.
Reading the drive output with a hand-held meter
A meter reading of about 280 V across a 230 V drive's output terminals does not mean the drive boosts voltage. The conversion 280 x 0.707 is not applicable; the sine-wave factor of 0.707 (and its inverse, 1.414) holds only for a sinusoid. 396 V (280 x 1.414) is a sine-equivalent peak that never appears on the wire. The true peak of the PWM pulse is the bus voltage, about 325 V at 230 V input, plus any reflection at the motor.
The meter reports a composite of fundamental and carrier-frequency content according to its own bandwidth. It reads neither the fundamental nor the pulse peak. Use it for a rough voltage-present check. Use a scope with adequate bandwidth for insulation-stress analysis.
Insulation failure mechanics: field strength, fatigue and moisture
Magnet wire insulation does not see the whole winding voltage. A turn in an inner layer touches only the adjacent few layers, slots hold insulating board, and coils in series usually have insulating material between overlapping sections. The stress on any one layer is set by the local voltage difference and the spacing, that is, by the field strength. Geometry therefore governs where failure starts: turn-to-turn at the first coil of a phase, where a fast edge distributes unevenly, and phase-to-ground at slot exits and end turns.
Insulation also fatigues under repeated electrical stress, much as metal does under cyclic load, and moisture accelerates the process. Partial discharge from repeated pulses erodes the insulation until a conductive path forms. A motor that has survived years on a VFD may have consumed part of its life, and a motor in a wet or humid location has less margin.
Symptoms versus causes
| Observation | Class of fault | Next measurement |
|---|---|---|
| Drive trips on ground fault at start, motor not hot, mechanical load normal | Electrical (insulation), not thermal | Insulation resistance phase-to-ground with the cable disconnected at the drive; then again with the motor leads disconnected to separate cable from motor. |
| Motor winding hot, healthy insulation resistance | Thermal (load, cooling, low-speed operation) | Motor current against nameplate, fan and airflow, operating speed range. |
| Repeated motor failures on 480 V with a long motor cable | Reflected peaks above corona inception voltage | Scope at motor terminals; cable length; motor stator insulation specification. |
| Drive input rectifier fails, motor healthy | Mains surge on the supply side | Supply quality; check surge protection ahead of the drive. |
| Standard motor on a 230 V VFD runs for years without a fault | Low stress case | None required; monitor insulation resistance during routine maintenance. |
Procedure for matching a motor to a VFD
- Record the supply voltage and compute the bus peak (VLL x 1.414).
- Record the motor cable length from drive to motor terminals.
- Check the nameplate and datasheet for an inverter duty marking and the manufacturer's stated spike withstand. Ask the manufacturer whether the stator passed a corona test, since that test defines the label.
- For a 230 V system with a short cable, a standard motor is a reasonable choice. For a 480 V system, or any long cable, specify an inverter duty motor or measure the peak at the motor terminals before committing.
- If the peak approaches the motor's withstand, reduce it at the source: shorten the cable, or add output filtering (an output reactor, dv/dt filter or sine filter) sized by the drive manufacturer for your cable length and drive rating.
- For a motor being rewound for VFD service, tell the shop before the rewind. Shops that stock inverter duty magnet wire and insulation systems as their standard can wind to the higher standard at no inventory penalty.
- Keep the existing motor if it is an older thick-insulation design in good condition, and confirm with insulation resistance readings and, if available, a shop partial discharge test.
Confirming the installation is within insulation limits
Record insulation resistance phase-to-ground at commissioning and at each maintenance interval, at the same temperature and the same test voltage, and trend it. A falling trend on a VFD-fed motor points to insulation erosion before a trip occurs. Scope the motor terminals once at commissioning and compare the measured peak and rise time against the motor's stated withstand. Confirm that the drive's ground-fault and overcurrent trips stay clear during full-speed-range operation. For motors in damp locations, add a check for condensation and keep the space heater or drain path functional.
What happens if I run a standard duty motor on a 230 V VFD?
Most standard motors run without insulation failure at this voltage, because a bus peak near 325 V leaves large margin against 600 V-rated magnet wire. Record cable length and track insulation resistance over time, since moisture and age reduce the margin.
What happens if I run a standard duty motor on a 480 V VFD with a long cable?
The bus peak is about 679 V, and cable resonance can raise the terminal peak well above that, which can exceed the corona inception voltage of a standard stator. Expect partial discharge erosion and eventual turn-to-turn or ground faults, so specify an inverter duty motor or add output filtering.
What happens if my meter reads 280 V on the output of a 230 V drive?
The reading is a meter artifact of the PWM waveform, not a voltage boost. The pulse peak is the DC bus (about 325 V at 230 V input) plus any reflection, so scope the motor terminals to see the stress the insulation actually carries.
What happens if mains spikes of 600 V hit a 240 V drive input?
The input rectifier and DC bus take the surge, not the motor, and a severe spike can fail the rectifier. Protect the supply side with surge suppression appropriate to the drive.
What happens if I cannot tell whether a motor is rated for inverter duty?
Stop guessing from the wire or the age of the motor and ask the motor manufacturer whether the stator was corona tested and what peak voltage and rise time it withstands. If the motor is on a 480 V drive with a long cable and you see repeated ground-fault trips or falling insulation resistance, escalate to the motor and drive manufacturers' official technical support with the measured terminal peak, cable length and insulation resistance trend.