IGBT (Insulated Gate Bipolar Transistor) is a minority carrier device with high input impedance and strong current carrying capacity. From the circuit designer’s point of view, IGBT has input characteristics of MOS devices. And the current output capability of the bipolar device is a voltage-controlled bipolar device. The purpose of the invention of the IGBT is to combine the advantages of the power MOSFET and the BJT device. It can be said that the IGBT is a combination of the power MOSFET and the BJT. The avatar of one. The advantages of both are concentrated in one and can have excellent performance.

IGBTs are suitable for many applications in power circuits, especially PWM drives, three-phase drives, which require high dynamic control and low noise. Other applications such as UPS, switching power supplies, etc. that require high switching frequency are also suitable for IGBTs. The IGBT is characterized by high dynamic performance, conversion efficiency, and low audible noise. It is also suitable for resonant mode inverter/inverter circuits. There are IGBTs optimized for low conduction loss and low switching loss. Device.

The main advantages of IGBTs for power MOSFETs and BJTs are as follows:

1. Has a very low on-voltage drop and excellent on-current density. Therefore, smaller size devices can be used to reduce costs.

2. Because the gate structure uses the same design of the MOS tube, the driving power is very small, and the driving circuit is also very simple. Compared with the thyristor/BJT current control type devices, the IGBT is very easy in high voltage and high current application scenarios. control.

3. It has better current conduction capability than BJT. The parameters are better in forward and reverse isolation.

In addition to its advantages, IGBTs have their shortcomings:
1. The switching speed is lower than the power MOSFET, but higher than BJT. Because it is a minority carrier device, the collector current residual causes the shutdown speed to be slower.

2. Because of the internal PNPN type thyristor structure, there is a certain probability that it will be locked.

The advantage of IGBTs is the ability to enhance voltage isolation. For example, for MOSFETs, as the breakdown voltage increases, the on-resistance increases very quickly because the thickness of the drift region and its own resistance must increase in order to increase the breakdown voltage. In practice, MOSFETs with high current carrying capacity and high breakdown voltage are generally not designed. For IGBTs, because of the highly concentrated injection of minority carriers during turn-on, the resistance of the drift region is greatly reduced. The forward voltage drop of the zone is only related to its thickness and is relatively independent of its own resistance.

basic structure
Figure 1 shows a simplified schematic of a typical N-channel IGBT fabricated using the DMOS process. This structure is just one of many possible configurations. It can be seen that in addition to the P+ implant layer, the silicon cross-over and vertical power MOSFET of the IGBT Basically the same. In the P-well of the gate region and the N+ source region, the IGBT has almost no difference from the MOSFET. The top N+ was the S pole or the emitter, and the bottom P+ was the D pole or the collector. If used during doping In the reverse order, then the P-channel IGBT. IGBT is fabricated because of the structure of the NPNP, so there will be a parasitic thyristor. It is generally not desirable to turn on the thyristor.


Figure 1 Typical N-channel IGBT structure

Some IGBTs are manufactured without the N+ buffer layer, which is called a non-through-type (NPT) IGBT. The opposite has a buffer layer called a through-type (PT) IGBT. If doped with this layer thickness Properly designed, this layer can greatly improve the performance of the entire device. Although the IGBT is similar to the MOSFET in the shape, the IGBT is more similar to BJT in practical work. This is because the drain layer (injection layer) of P+ can minority carriers. The conduction modulation characteristics caused by the injection of the N-drift region. 

Figure 2 IGBT equivalent circuit

From the above analysis, the equivalent circuit diagram of the IGBT can be drawn (Fig. 2). The equivalent circuit includes MOSFET, JFET, NPN and PNP transistors. The collector of PNP is connected to the base of NPN. The collector of NPN passes through the base of JFET and PNP. The poles are connected. NPN and PNP represent parasitic thyristors. This thyristor will bring a regenerative feedback loop. RB is the NPN BE junction resistor, which is to ensure that the parasitic thyristor is not locked to ensure the IGBT. Not locked. The JFET represents the contraction current between any two adjacent IGBTs. The JFET is present in most voltage ranges, leaving the MOSFET at a low voltage resulting in a low RDS(on) value. Figure 3 shows the IGBT. Circuit symbol. The three poles are called collector (C), gate (G) and emitter (E).

Figure 3 IGBT circuit symbol

IXYS products include both NPT and PT type IGBTs. The two types of physical structures are shown in Figure 4. As mentioned earlier, the PT type has an additional layer. This has two main functions: (i) avoid because of The depletion region is extended by the high voltage, thereby avoiding the punch-through type failure. (ii) Since the hole portion injected in the P+ collector region is recombined at this layer, the residual current at the turn-off is reduced, thereby shortening the off Breakdown time. NPT type IGBTs have the same forward and reverse breakdown voltages and are suitable for AC applications. PT type IGBTs have a reverse breakdown voltage lower than the forward breakdown voltage and are suitable for DC circuits ( Because the device in the DC circuit does not need to reverse the voltage).


Figure 4 NPT and PT type IGBT structure

Table 1: Comparison of IGBT characteristics between NPT and PT

Operating mode
Positive turn-off and conduction mode

As shown in Figure 1, when the collector-emitter plus forward voltage and the gate and emitter are shorted, the IGBT enters the positive shutdown mode. At this point, J1 and J3 are forward biased and J2 is reverse biased. The depletion region at both ends of J2 partially diffuses to the P base and N drift regions.

When the short circuit between the gate and the emitter is removed, and a sufficient voltage is applied to the gate to reverse the silicon in the P base region, the IGBT is transferred from the forward turn-off mode to the forward conduction mode. In the mode, a conduction channel is formed between the N+ emitter and the N-drift region. The electrons of the N+ emitter flow to the N-drift region through the channel. The electrons flowing to the N-drift region reduce the potential of the N-drift region. The junction of the P+ collector/N-drift region is forward biased so that high-density minority carrier holes are injected from the P+ collector into the N-drift region. When the injected carrier density is much higher than the background density In the case of the N-drift region, a condition called hole ion current is established. This hole ion current attracts electrons from the emitter to the emitter to maintain local charge neutralization. Thus, the N-drift region is established. The concentration of certain holes and electrons is concentrated. This type of partitioning greatly improves the conductivity of the N-drift region. This mechanism is called the conduction modulation of the N-drift region.

Reverse shutdown mode

When a negative voltage is applied between the collector and the emitter as shown in Figure 1, J1 is reverse biased, and its depletion region is diffused to the N-drift region. The breakdown voltage for reverse turn-off is made by P+ collector/N. – The open base BJT formed by the drift region / P base is determined. If the N-drift region is insufficiently doped, the device will be easily broken down. To obtain the required breakdown voltage, the N-drift region must be controlled. Resistance and thickness.

To obtain specific parameters for reverse breakdown voltage and forward voltage drop, the following is the formula for calculating the width of the N-drift region:

among them:

LP: minority carrier spur length

Vm: maximum shutdown voltage

Εo: dielectric constant of free region

Εs: dielectric constant of silicon

q: charge

ND: Doping density of the N drift region

Note: Reverse turn-off of IGBTs is rare in most applications, but the use of anti-parallel diodes (FRED) is common.

Output characteristics
Figure 5 shows the forward output characteristic of an NPT-IGBT. This is a group of curves, each representing a different gate-emitter voltage. The collector current (IC) is a function of VCE when VGE is fixed. .


Figure 5 I-V output curve of NPT-IGBT

Note that the offset voltage is 0.7V. This is because there is an additional PN junction for the IGBT of the P+ collector. This PN junction distinguishes the characteristics of the IGBT from the MOSFET.

Transmission characteristics
The transmission characteristics refer to the response function of ICE to VGE changes at different temperatures, such as 25 degrees, 125 degrees, and -40 degrees. As shown in Figure 6, the gradient of the transmission characteristics at a given temperature is called the device. Transconductance (gfs) at temperature.


Figure 6 IGBT transmission characteristics

Generally speaking, it is necessary to obtain a high current capability at a lower gate voltage. It is desirable that the value of gfs is relatively large. The structure of the channel and the gate determines the value of gfs. Both gfs and RDS(on) are controlled by the length of the channel, and The length of the channel is determined by the difference between the diffusion depth of the P-base and the N+ emitter. The tangent on the transmission characteristic determines the threshold/threshold voltage (VGE(th)) of the device.


Figure 7 Transconductance characteristics of an IGBT

Figure 7 shows the transconductance characteristics (IC-gfs) of an IGBT. As the collector current increases, gfs increases, but as the collector current continues to increase, the growth curve of gfs slowly and slowly. This is because of parasitic MOSFETs. The saturation phenomenon slows down the increase of the driving current of the base of the PNP transistor.

Switching characteristics
The switching characteristics of the IGBT are very similar to those of the MOSFET. The main difference is that since the N-drift region stores charge, it will cause a residual collector current. This residual current increases the turn-off loss and requires two devices in the half-bridge circuit. The dead time between breaks increases accordingly. Figure 8 shows the test circuit for switching characteristics. Figure 9 shows the corresponding voltage and current waveforms for turn-on and turn-off. IXYS IGBT products use a gate voltage of 15V to 0V during testing. In order to reduce the switching loss, it is recommended to add a negative voltage (such as -15V) to the gate when it is turned off.

Figure 8 switch characteristic test circuit

The switching speed of the IGBT is limited by the lifetime of minority carriers in the N-drift region of the base of the parasitic PNP transistor. This region is inoperable externally, so there is no external means to increase the rate at which this charge is removed. To increase the switching speed. The only way to remove this charge is to re-neutralize inside the IGBT. In addition, adding N+ buffer to collect minority carrier charge can increase the neutralization speed of this charge.

Figure 9 IGBT turn-off voltage and current waveform

Eon represents the conduction energy, which is the integral of IC*VCE in the VCE range from 10% ICE to 90%. The amount of conduction energy depends on the reverse recovery characteristics of the freewheeling diode, so if the IGBT contains a freewheeling diode Be sure to pay special attention.

Eoff represents the turn-off energy, which is the integral of IC*VCE in the 10% VCE to 90% IC interval. Eoff is the main component of the IGBT switching loss.

Lock/up (Latch-up)
In the on state, the internal current of the IGBT is as shown in Fig. 10. The holes injected into the N-drift region from the P+ collector form two current paths. A part of the holes disappear due to electron neutralization with the MOSFET channel. Part of the hole is attracted to the vicinity of the reverse layer by the negative charge of the electron. From the epitaxy through the P layer, a voltage drop is formed in the body ohmic resistance region. If this voltage is large enough, the N+P junction will be forward biased. At the same time, a large amount of electrons are injected from the emitter and the parasitic NPN transistor will be turned on. If this phenomenon occurs, the parasitic NPN and PNP transistors will be turned on at the same time, so the thyristor composed of the two tubes will be latched up. Thus, the entire IGBT is locked. Once the lock occurs, the gate voltage will lose control of the current of the collector. The only way to turn off the IGBT is to force the commutation, just like in a real thyristor. .

Figure 10 Current flow in the IGBT on state

If this locked state cannot be terminated quickly, the IGBT will be burnt due to excessive power dissipation. The maximum peak current that the IGBT can pass without locking is called (ICM). The device data sheet will indicate This parameter. Exceeding this current value, a sufficiently large peripheral voltage drop activates the thyristor and causes locking.

Safe Operating Area (SOA)
The so-called safe working area refers to a range of current-voltage, in which the device can work safely without being damaged. For IGBT, this interval consists of the largest collector-emitter voltage VCE and collector current. Ic defines that the IGBT can safely operate without damage in this interval. The safe working area of ​​the IGBT has the following types: forward biased safe working area (FBSOA), reverse biased safe working area (RBSOA) and short circuit safe working area. (SCSOA).

Forward Biased Safe Work Area (FBSOA)
For inductive load applications, FBSOA is an important feature. It is determined by the maximum collector-emitter voltage and saturated collector current. In this mode, electrons and holes move through the drift region and remain relatively high. The collector voltage. The relationship between the density of electrons and holes in the drift region and the current current density is:

Where Vsat,n and Vsat,p are the saturation drift speeds of electrons and holes, respectively. The net positive charge of the drift region is:

This charge determines the electric field distribution in the drift region. Under steady-state positive shutdown conditions, the charge in the drift region is equal to ND. In the positive safe operating range, the net charge is much larger than ND because of the density of holes. Far greater than the density of the electron flow.

Reverse Bias Safe Work Area (RBSOA)
For turn-off transient analysis, RBSOA is an important state. The current that can be turned off is limited to twice the rated current of the IGBT. For example, an IGBT with a rated current of 1200A can turn off the maximum current is 2400A. The maximum current is The peak voltage of the collector and emitter when turned off. The peak value of VCE is equal to the product of DC voltage and LбdIC/dt. Lσ is the stray inductance of the power circuit. The relationship between the maximum current IC and VCE under RBSOA is shown in Figure 11. .

Figure 11 Reverse safety working area of ​​IGBT

In this mode, the bias of the gate is 0 or a negative voltage, so that the current in the drift region is only through the holes (N-channel IGBT). The holes increase the charge in the drift region, so the P-base/N drift The electric field of the node is increased. The net charge of the empty charge region under this condition is:

Where Jc is the total collector current. The avalanche voltage of RBSOA is:

Short Circuit Safe Work Area (SCSOA)
For devices operating in motor control applications, a key requirement is to be able to safely turn off when the load is shorted. When the current is overloaded, the current of the collector rises rapidly until the device can withstand the limits. The device can not be damaged under such conditions. It is possible to limit the current amplitude to a safe level before the control circuit detects the short-circuit condition and turns off the device.

The collector current IC of the IGBT is a function of the gate-emitter voltage VGE and the temperature T. The transmission characteristics shown in Figure 6 indicate the maximum IC value for a given VGE. For a VGE of 15V, the value is limited to 80A, approximately It is 1.5 times the rated value. Considering that the short-circuit current is often 6-7 times of the rated current, this value is very small.

Figure 12 SCSOA test circuit

Figure 12 shows a test circuit for SCSOA. The short-circuit inductor value determines the operating mode of the circuit. When this value is uH, the circuit operates in a similar manner to the normal inductive load switch. When the IGBT is turned on, VCE drops to the saturation voltage. The rate of dIC/dt increases and the IGBT gradually saturates. When the collector current is higher than 2 times the rated current, the shutdown operation is not allowed because it is beyond the RBSOA. If the short circuit occurs, it must wait for the device to reach the active working area. Turn off the IGBT within 10us to prevent the device from being damaged due to overheating.

IGBT Module Application: Fruit Labeling Machine

Source: https://www.slw-ele.com/igbt-knowledge-3.html