Insulated-gate bipolar transistors (IGBTs) are power devices that blend MOSFET and BJT features. They offer high input impedance, voltage control, and excellent current handling. IGBTs are key in power electronics for motor drives and inverters.

IGBTs have a four-layer structure with a MOSFET gate on top. The MOSFET controls conductivity, while the BJT handles high currents. This combo allows IGBTs to outperform MOSFETs and BJTs in high-voltage, high-current applications.

Structure of IGBTs

• IGBTs are power semiconductor devices that combine the high input impedance and gate control of a MOSFET with the high current and low saturation voltage capability of a bipolar junction transistor (BJT)
• The structure of an IGBT consists of four alternating layers (P-N-P-N) that form a thyristor-like structure with a MOSFET gate for controlling conductivity
• IGBTs are widely used in power electronics applications such as motor drives, inverters, and switch-mode power supplies due to their superior performance compared to MOSFETs and BJTs in high-voltage, high-current applications

Vertical cross-section

• The vertical cross-section of an IGBT reveals its four-layer structure (P-N-P-N) with a MOSFET gate on top
• The topmost layer is the emitter (P+), followed by the body region (N-), drift region (N-), and collector (P+) at the bottom
• The MOSFET gate is formed by the polysilicon gate electrode, gate oxide, and the body region (N-) underneath
• The drift region (N-) is lightly doped to support high blocking voltages in the off-state
• The collector (P+) layer at the bottom provides a low-resistance path for the hole current during the on-state

Equivalent circuit representation

• The equivalent circuit of an IGBT consists of a MOSFET in series with a wide-base PNP BJT
• The MOSFET controls the base current of the PNP BJT, which in turn controls the collector-emitter current
• The MOSFET gate terminal is the input, while the collector and emitter terminals of the PNP BJT are the output
• The MOSFET drain is connected to the base of the PNP BJT, forming a Darlington-like configuration
• The equivalent circuit representation helps in understanding the operation and modeling of IGBTs

Operating principle of IGBTs

• The operating principle of IGBTs is based on the combined action of the MOSFET and the PNP BJT
• The MOSFET gate controls the conductivity of the device, while the PNP BJT handles the high current conduction
• When a positive gate voltage is applied, the MOSFET creates an inversion layer (channel) in the body region, allowing electrons to flow from the emitter to the collector
• The electron current flowing through the drift region acts as the base current for the PNP BJT, causing it to turn on and conduct hole current from the collector to the emitter

Controlling conductivity via gate voltage

• The conductivity of an IGBT is controlled by the voltage applied to the MOSFET gate
• When the gate voltage exceeds the threshold voltage (typically 5-7 V), the MOSFET creates an inversion layer (channel) in the body region, allowing electron current to flow
• The electron current flowing through the drift region acts as the base current for the PNP BJT, causing it to turn on and conduct hole current
• As the gate voltage increases, the channel conductivity increases, leading to higher electron and hole currents
• The gate voltage can be used to control the IGBT in linear (proportional) or switched (on/off) modes

On-state characteristics

• In the on-state, the IGBT conducts high current with a low voltage drop across the collector-emitter terminals
• The on-state voltage drop consists of the MOSFET channel resistance, drift region resistance, and the PNP BJT saturation voltage
• The on-state voltage drop is typically higher than that of a MOSFET but lower than that of a BJT
• The on-state current is limited by the MOSFET channel resistance and the PNP BJT gain
• The on-state characteristics can be improved by optimizing the device geometry, doping profiles, and carrier lifetime

Off-state characteristics

• In the off-state, the IGBT blocks high voltage across the collector-emitter terminals with minimal leakage current
• The off-state blocking capability is determined by the thickness and doping of the drift region (N-)
• The electric field distribution in the drift region is shaped by the field-stop layer (N+) and the collector (P+) to achieve high blocking voltages
• The leakage current in the off-state is mainly due to the thermal generation of carriers in the drift region and the surface leakage at the junction terminations
• The off-state characteristics can be improved by optimizing the drift region design, field-stop layer, and junction termination techniques

Comparison of IGBTs vs MOSFETs

• IGBTs and MOSFETs are both voltage-controlled power semiconductor devices used in power electronics applications
• The choice between IGBTs and MOSFETs depends on the specific application requirements such as voltage and current ratings, switching speed, and efficiency
• IGBTs are generally preferred for high-voltage (>1 kV), high-current applications, while MOSFETs are better suited for lower-voltage, high-frequency applications

Voltage and current ratings

• IGBTs have higher voltage ratings compared to MOSFETs due to their thyristor-like structure and thick drift region
• Typical IGBT voltage ratings range from 600 V to 6.5 kV, while MOSFET voltage ratings are usually below 1 kV
• IGBTs can handle higher current densities than MOSFETs due to the bipolar conduction mechanism (electrons and holes)
• IGBT current ratings can reach several hundred amperes, while MOSFET current ratings are typically lower

Switching speed and losses

• MOSFETs have faster switching speeds compared to IGBTs due to their unipolar conduction mechanism (electrons only)
• MOSFET switching times are in the range of tens to hundreds of nanoseconds, while IGBT switching times are typically a few microseconds
• IGBTs have higher switching losses than MOSFETs due to the slower turn-off process and the tail current caused by the minority carrier (hole) storage
• IGBT switching losses increase with the switching frequency, limiting their usage in high-frequency applications

Applications and trade-offs

• IGBTs are widely used in medium to high-power applications such as motor drives, inverters, and switch-mode power supplies
• MOSFETs are preferred in low to medium-power, high-frequency applications such as switched-mode power supplies, DC-DC converters, and RF power amplifiers
• The choice between IGBTs and MOSFETs involves a trade-off between the voltage and current handling capability, switching speed, and efficiency
• In applications where high voltage and current handling are critical, IGBTs are the preferred choice, while MOSFETs are better suited for applications demanding high switching speeds and efficiency

IGBT turn-on process

• The IGBT turn-on process involves the charging of the device capacitances and the establishment of the on-state current
• The turn-on process is initiated by applying a positive gate voltage, which creates an inversion layer (channel) in the body region
• The turn-on process can be divided into several stages, each characterized by specific capacitance charging and current dynamics

Capacitances and charges

• The main capacitances involved in the IGBT turn-on process are the gate-emitter capacitance (Cge), gate-collector capacitance (Cgc), and collector-emitter capacitance (Cce)
• Cge is the sum of the gate oxide capacitance and the capacitance between the gate and the inversion layer
• Cgc is the capacitance between the gate and the drift region, which varies with the depletion width in the drift region
• Cce is the capacitance between the collector and emitter terminals, which includes the depletion capacitance of the drift region and the diffusion capacitance of the PNP BJT
• During the turn-on process, these capacitances are charged by the gate driver circuit, influencing the turn-on speed and losses

Stages of turn-on transient

• The IGBT turn-on transient can be divided into four main stages:
  1. Gate charging: The gate-emitter capacitance (Cge) is charged until the gate voltage reaches the threshold voltage (Vth)
  2. Miller plateau: The gate voltage remains constant at the Miller voltage (Vm) while the gate-collector capacitance (Cgc) is charged, and the collector voltage starts to fall
  3. Current rise: The collector current increases rapidly as the MOSFET channel becomes fully enhanced, and the PNP BJT starts to conduct
  4. Voltage fall: The collector voltage falls to the on-state value, and the collector current reaches its steady-state value
• Each stage has its own time constant and power loss contribution, which can be optimized by proper gate driver design and device selection

Modeling turn-on delay and rise time

• The turn-on delay (td(on)) is the time from the application of the gate voltage to the start of the collector current rise
• The turn-on delay can be modeled using the gate-emitter capacitance (Cge) and the gate resistance (Rg) as: $t_{d(on)} = R_g C_{ge} \ln(\frac{V_{gg} - V_{th}}{V_{gg} - V_m})$
• The current rise time (tr) is the time from the start of the collector current rise to the point where it reaches 90% of its steady-state value
• The current rise time can be modeled using the gate-collector capacitance (Cgc) and the gate resistance (Rg) as: $t_r = 2.2 R_g C_{gc}$
• Accurate modeling of the turn-on delay and rise time is essential for optimizing the gate driver design and minimizing the turn-on losses

IGBT turn-off process

• The IGBT turn-off process involves the discharging of the device capacitances and the removal of the excess carriers from the drift region
• The turn-off process is initiated by applying a negative gate voltage, which removes the inversion layer (channel) in the body region
• The turn-off process can be divided into several stages, each characterized by specific capacitance discharging and current dynamics

Capacitances and charges

• The main capacitances involved in the IGBT turn-off process are the same as in the turn-on process: gate-emitter capacitance (Cge), gate-collector capacitance (Cgc), and collector-emitter capacitance (Cce)
• During the turn-off process, these capacitances are discharged by the gate driver circuit, influencing the turn-off speed and losses
• The excess carriers (electrons and holes) stored in the drift region during the on-state must be removed during the turn-off process, contributing to the turn-off delay and losses

Stages of turn-off transient

• The IGBT turn-off transient can be divided into four main stages:
  1. Gate discharging: The gate-emitter capacitance (Cge) is discharged until the gate voltage reaches the Miller voltage (Vm)
  2. Voltage rise: The collector voltage rises rapidly as the MOSFET channel becomes pinched off, and the PNP BJT starts to turn off
  3. Current fall: The collector current decreases rapidly as the excess carriers are removed from the drift region
  4. Tail current: The collector current exhibits a slow decay (tail current) due to the residual excess carriers in the drift region
• Each stage has its own time constant and power loss contribution, which can be optimized by proper gate driver design and device selection

Modeling turn-off delay and fall time

• The turn-off delay (td(off)) is the time from the application of the negative gate voltage to the start of the collector voltage rise
• The turn-off delay can be modeled using the gate-emitter capacitance (Cge) and the gate resistance (Rg) as: $t_{d(off)} = R_g C_{ge} \ln(\frac{V_{gg} - V_m}{V_{gg} - V_{th}})$
• The voltage rise time (tv) is the time from the start of the collector voltage rise to the point where it reaches 90% of its off-state value
• The voltage rise time can be modeled using the gate-collector capacitance (Cgc) and the gate resistance (Rg) as: $t_v = 2.2 R_g C_{gc}$
• The current fall time (tf) is the time from the start of the collector current fall to the point where it reaches 10% of its on-state value
• The current fall time can be modeled using the excess carrier lifetime (τ) and the drift region width (Wd) as: $t_f = \frac{W_d^2}{2 D_n}$, where Dn is the electron diffusion coefficient
• Accurate modeling of the turn-off delay, voltage rise time, and current fall time is essential for optimizing the gate driver design and minimizing the turn-off losses

Safe operating area of IGBTs

• The safe operating area (SOA) of an IGBT defines the range of voltage and current conditions under which the device can operate without damage or degradation
• The SOA is determined by the device's physical limitations, such as the maximum junction temperature, the breakdown voltage, and the current handling capability
• The SOA can be divided into three main regions: forward-bias SOA, reverse-bias SOA, and switching SOA

Forward-bias SOA

• The forward-bias SOA defines the maximum collector current and collector-emitter voltage that the IGBT can sustain in the on-state
• The forward-bias SOA is limited by the maximum junction temperature, which is determined by the power dissipation and the thermal resistance of the device
• The forward-bias SOA can be extended by using a larger chip size, improving the thermal management, or using parallel-connected devices
• The forward-bias SOA is typically specified in the device datasheet as a graph of the collector current versus the collector-emitter voltage

Reverse-bias SOA

• The reverse-bias SOA defines the maximum collector-emitter voltage that the IGBT can block in the off-state without experiencing avalanche breakdown
• The reverse-bias SOA is limited by the breakdown voltage of the device, which is determined by the doping and thickness of the drift region
• The reverse-bias SOA can be extended by using a thicker drift region, optimizing the field-stop layer, or using junction termination techniques
• The reverse-bias SOA is typically specified in the device datasheet as a maximum collector-emitter voltage rating

Switching SOA

• The switching SOA defines the maximum collector current and collector-emitter voltage that the IGBT can withstand during the switching transients
• The switching SOA is limited by the dynamic avalanche breakdown, which occurs when the device is subjected to high voltage and high current simultaneously
• The switching SOA can be extended by using a softer switching technique (e.g., zero-voltage or zero-current switching), optimizing the gate driver design, or using a snubber circuit
• The switching SOA is typically specified in the device datasheet as a graph of the collector current versus the collector-emitter voltage, with different curves for various gate resistances and junction temperatures

IGBT gate drive requirements

• The gate drive requirements for an IGBT are critical for ensuring proper operation, optimizing the switching performance, and protecting the device from damage
• The gate driver circuit must provide the necessary voltage and current to charge and discharge the IGBT's input capacitance within the desired switching time
• The gate driver must also provide isolation between the control and power circuits, as well as protection against over-voltage, over-current, and short-circuit conditions

Gate voltage and current

• The gate voltage required to turn on an IGBT is typically in the range of 10-20 V, while the turn-off voltage is usually between -5 and -15 V
• The gate voltage must be higher than the threshold voltage (Vth) to ensure complete turn-on and lower than the maximum gate-emitter voltage rating to avoid gate oxide damage
• The gate current required to charge and discharge the input capacitance depends on the desired switching speed and the gate resistance
• The peak gate current can reach several amperes for fast switching applications, requiring a low-impedance gate driver circuit

Gate driver circuit topologies

• There are several gate driver circuit topologies used for IGBTs, depending on the application requirements and the isolation level needed
• The simplest topology is the direct-coupled gate driver, which uses a low-impedance voltage source (e.g., a bipolar power supply) to drive the IGBT gate directly
• The isolated gate driver topologies use a transformer or an optocoupler to provide galvanic isolation between the control and power circuits
• The transformer-isolated gate driver uses a pulse transformer to transmit the gate signal across the isolation barrier, while the optocoupler-isolated gate driver uses an LED and a photodetector
• The isolated gate driver topologies require additional circuitry, such as a DC-DC converter or a charge pump, to generate the isolated gate voltage

Protection and isolation

• The gate driver circuit must provide protection against various fault conditions to ensure the safe operation of the IGBT
• The over-voltage protection limits the gate-emitter voltage to a safe level, typically using a Zener diode or a transient voltage suppressor (TVS)
• The over-current protection limits the collector current to a safe level, typically using a desaturation detection circuit or a current transformer
• The short-circuit protection detects and turns off the IGBT within a few microseconds in case of a short-circuit