Posted on 06 February 2020

Types of Faults for IGBT and MOSFET Modules








Power semiconductors have to be protected from non-permissible stress in every operational state. Leaving SOAs (Safe Operating Areas) leads to damage and therefore reduces component's life. In the worst case scenario, the component will be immediately destroyed. This is why it is important to detect critical states and faults and respond to them with suitable measures. The explanations given here refer mainly to IGBTs, but may also be applied to power MOSFETs in the same manner. Any matters relevant to MOSFETs in particular will be pointed out separately.

Fault currents

Fault currents are collector/drain currents that exceed standard operating values of a certain application due to control or load errors. They might lead to damage to the power semiconductors due to the following mechanisms:

  • thermal destruction caused by high power dissipation
  • dynamic avalanche
  • static or dynamic latch-up
  • overvoltages that occur in connection with fault currents

A distinction is made between the following fault currents:



  • usually relatively low collector current di/dt (depending on load inductance and driving voltage)
  • fault current is conducted through the DC link
  • transistor does not desaturate


  • reduced load impedance
  • converter control error

Short-circuit current


  • very steep collector current di/dt
  • fault current is conducted through the DC link
  • transistor is desaturated


  • arm short circuit
  • by defective switch
  • by faulty driver pulses for the switches
  • load short circuit
  • by faulty isolation
  • human errors (wrong connection wiring, etc.)

Earth fault current (Case 3 in figure 1)


  • collector current di/dt is dependent on earth inductance and driving voltage
  • earth fault circuit is not closed over DC link
  • desaturation of the transistor is dependent on fault current value


  • connection between a live conductor and earth potential (caused by faulty isolation or human error)

Causes of fault currents

Figure 1. Causes of fault currents


Dangerous overvoltages occur if the break-down voltages of power semiconductors are exceeded. This applies to both transistors and diodes. With respect to IGBTs and MOSFETs, overvoltages may occur between collector and emitter (or drain and source) - i.e. between the main terminals - as well as between gate and emitter (or gate and source) - i.e. between the control terminals.

Causes of overvoltages between main terminals

Figure 2 shows basic types of overvoltages between the main terminals of power semiconductors on the example of a commutation circuit.

Types of overvoltages in a commutation circuit

Figure 2. Types of overvoltages

Generally speaking, in commutation circuits, a distinction is made between external and internal overvoltages. In connection with this, an "external overvoltage" is to be seen as a transient increase in the impressed commutation voltage vK. This may happen, for example, in the DC voltage mains in electric traction power circuits or in any power supply system. Increased DC link voltages are to be seen in the same way (caused, e.g., by active feedback loads or control errors in pulse rectifiers).

"Internal overvoltages" are generated, for example, if the power electronic switch is turned off against the commutation circuit inductance LK (Δv = LK·diK/dt) or if oscillations occur due to switching procedures. The following cases are typical examples of the generation of switching overvoltages:

  • Active turn-off of load current iL by the active elements of switches S1 and S2 during normal converter operation: In many SMPS applications (Switch Mode Power Supply), the inductance LK is generated as a result of the stray inductance of transformers, which may be as much as 1-100 μH
  • Reverse recovery di/dt during passive turn-off of fast diodes in hard switching converters or ZCS (Zero Current Switching) converters: Owing to their operating principle, ZCS converters may also show an increased commutation inductance within the range of 10 μH
  • High di/dt (...10 kA/μs...) in the event of short circuits and during turn-off of short circuit currents in converters with DC voltage link
  • Active interruption of DC link currents in CSI topologies (failure)

Furthermore, overvoltages in power electronic devices may be generated by static or dynamic asymmetries in switches connected in series. Overvoltages during normal operation of converters and converter fault operation may appear as periodic (...Hz...kHz...) or aperiodic overvoltages.

Causes of overvoltages between control terminals

Overvoltages between control terminals of IGBTs and MOSFETs can be due to:

  • supply voltage error in the driver stage
  • dv/dt feedback (displacement current to the gate) via Miller capacitance
  • emitter/source di/dt feedback
  • increase in gate voltage during active clamping
  • parasitic oscillations in the gate circuit (e.g. connection to collector/drain, transient oscillations between gate circuits of parallelled transistors, etc.)


Dangerous overtemperatures occur if the maximum junction temperature specified by the device manufacturer is exceeded (e.g. Tjmax = 150°C...175 °C for silicon devices).

During converter operation, overtemperatures might be generated by:

  • an increase in energy dissipation caused by fault currents
  • an increase in energy dissipation caused by defective drivers
  • failure or malfunction of the cooling system


For more information, please read:

Overload and Short Circuit Behavior of IGBTs and MOSFETs

Overvoltage Limitation for Power Transistors

Interruption of DC Fault Currents

High Voltage Power Thyristor with Built-in Protective Elements


VN:F [1.9.17_1161]
Rating: 6.0/6 (2 votes cast)
Types of Faults for IGBT and MOSFET Modules, 6.0 out of 6 based on 2 ratings

This post was written by:

- who has written 197 posts on PowerGuru - Power Electronics Information Portal.

Contact the author

Leave a Response

You must be logged in to post a comment.