Insulated Gate Bipolar Transistor | IGBT

IGBT combines the best features of the two devices that dominated power electronics before it. On one side stood the current-controlled Power BJT, with excellent conduction characteristics but slow switching, low input impedance and a risk of secondary breakdown. On the other stood the voltage-controlled Power MOSFETs, with fast switching and high input impedance but poor conduction characteristics and a problematic parasitic diode at higher ratings. Because their unipolar structure raises ON-state resistance as the voltage rating increases, Power MOSFETs also become lossy at high voltage.

Engineers therefore needed a device that combined the strengths of both. The IGBT was developed in the early 1980s and quickly became popular among engineers for its superior characteristics. It has MOSFET-like input characteristics and Power BJT-like output characteristics, and its three terminals are Gate, Collector and Emitter.

The IGBT is also known under several other names, such as Metal Oxide Insulated Gate Transistor (MOSIGT), Gain Modulated Field Effect Transistor (GEMFET), Conductively Modulated Field Effect Transistor (COMFET) and Insulated Gate Transistor (IGT).

Structure of IGBT

The structure of IGBT closely resembles that of the Power MOSFET except for one extra layer known as the injection layer, which is p⁺ unlike the n⁺ substrate in the MOSFET. This injection layer is the key to the superior characteristics of the IGBT, because it injects holes into the drift region and conductivity-modulates it. The other layers are called the drift and body regions, and their junctions are labeled J₁ and J₂ in the figure below, which shows an n-channel IGBT.

Closer observation of the structure reveals an n-channel MOSFET together with two BJTs, Q₁ and Q₂, as shown in the figure. Q₁ is a p⁺n^–p BJT and Q₂ is n^–pn ⁺ BJT. R_d is the resistance offered by the drift region and R_b is the resistance offered by the p body region. The collector of Q₁ forms the base of Q₂, and the collector of Q₂ forms the base of Q₁, so the equivalent circuit model of the IGBT emerges as shown in the figure below.

This back-to-back connection of the two transistors forms a parasitic thyristor, also visible in the figure above.

An n-channel IGBT turns ON when the collector sits at a positive potential with respect to the emitter and the gate is driven to a sufficiently positive potential (>V_GET). These conditions create an inversion layer just below the gate, the channel forms, and current begins to flow from collector to emitter.

The collector current I_c in an IGBT consists of two components, I_e and I_h. I_e is the current carried by electrons flowing from collector to emitter through the injection layer, the drift layer and finally the formed channel. I_h is the hole current flowing from collector to emitter through Q₁ and the body resistance R_b. Hence

I_h is only a small fraction of the total current, so I_c ≈ I_e.
A phenomenon called latching up occurs when the collector current exceeds a threshold (I_CE). The parasitic thyristor then turns on, and the gate terminal loses control over the collector current, so the device cannot be turned off even if the gate voltage drops below V_GET. Turning the IGBT off then requires commutation circuits similar to the forced commutation of thyristors, and an uncontrolled latched device may be damaged. Modern cell designs largely suppress this static latch-up, but it remains a consideration at high currents and temperatures.

Characteristics of IGBT

Static I-V Characteristics of IGBT

The figure below shows the static i-v characteristics of an n-channel IGBT together with a circuit diagram marking the parameters.

The graph resembles that of a BJT, except that the parameter held constant for each plot is V_GE, because the IGBT is a voltage-controlled device unlike the current-controlled BJT. In the OFF state (V_CE positive and V_GE < V_GET), junction J₂ blocks the forward voltage, and under reverse bias, when V_CE is negative, J₁ blocks the voltage.

Transfer Characteristics of IGBT

The transfer characteristic of the IGBT, shown below, is identical in shape to that of the Power MOSFET. The device stays OFF until V_GE exceeds a threshold value V_GET, after which it enters the ON-state.

Switching Characteristics of IGBT

The figure below shows the typical switching characteristic of IGBT.

Turn-on time (t_on) consists of two parts: delay time (t_dn) and rise time (t_r). During the delay time, the collector current rises from the leakage value (I_CE) to 0.1 IC while the collector-emitter voltage falls from V_CE to 0.9V_CE. During the rise time, the collector current increases from 0.1 I_C to I_C while the collector-emitter voltage falls from 0.9 V_CE to 0.1 V_CE.

The turn-off time t_off consists of three components: delay time (t_df), initial fall time (t_f1) and final fall time (t_f2). The delay time is the interval in which the collector current falls from I_C to 0.9 I_C and V_CE begins to rise. During the initial fall time, the collector current falls from 0.9 I_C to 0.2 I_C and the collector-emitter voltage rises to 0.1 V_CE. The final fall time is the interval in which the collector current falls from 0.2 I_C to 0.1 I_C while 0.1V_CE rises to its final value V_CE.

Advantages and Disadvantages of IGBT

Advantages:-
The main advantages of IGBT are listed below

• Lower gate drive requirements
• Low switching losses
• Small snubber circuitry requirements
• High input impedance
• Voltage controlled device
• Bipolar conductivity modulation gives a lower On-state voltage drop and power loss than a Power MOSFET of similar rating.
• Enhanced conduction due to bipolar nature
• Better Safe Operating Area

Disadvantages:-
The main disadvantages of IGBT are listed below

• Cost
• Latching-up problem
• High turn off time compared to PMOSFET