Case · gate branch

Case: IGBT Failed Again Because One Gate Branch Was Weak

After IGBT replacement, the same position failed again. Unloaded branch comparison found a weak gate path caused by an open resistor/connection.

Case summary

After IGBT replacement, the same position failed again. Unloaded branch comparison found a weak gate path caused by an open resistor/connection.

This is a reusable diagnostic pattern; exact values and reference designators remain board-specific.

Initial symptom

Power-off evidence

  1. DC bus discharged and no residual short remained.
  2. Gate resistors, clamps and connectors were compared.
  3. The controller was not condemned only because PWM was missing.

Controlled live evidence

PointReferenceObserved pattern
PWM outputsController groundBoth channels coherent
Driver primaryPrimary driver returnInputs coherent
Affected secondaryMatching isolated returnWeaker than equivalent branches
Gate-emitterThat IGBT emitterSlow turn-off / reduced amplitude

Diagnosis

Controller output and driver-primary pulses were correct. One isolated secondary and its gate-emitter waveform had lower amplitude and slow turn-off. The fault was local to that branch, not the controller.

Repair action

The open gate resistor/connection and parts damaged by the earlier short were repaired. Every equivalent branch was verified before the power devices were installed.

Verification

  1. Controller and driver verified with the power stage disconnected.
  2. Equivalent branches confirmed.
  3. First bus power-up current-limited and unloaded.
  4. Short load test completed without abnormal protection or heating.

Stop conditions

Technical sources

Related repair map

Follow the evidence path from symptom to measurement, circuit and repair case.