Application of power devices in static frequency conversion technology

Due to the influence of power electronic device technology, the static frequency conversion technology in my country before the 1980s has been in a state of stagnation. The medium-frequency power supply for induction heating in various industries basically uses medium-frequency frequency conversion motors for power supply. With the development of power electronic devices in the early 1990s, the domestic medium-frequency motor groups are currently being eliminated, and static frequency conversion equipment has begun to be used in large quantities, especially in the audio and super audio fields. In the 1980s and 1990s, domestic static frequency conversion basically used thyristors as power switching elements, and the process level was basically capped at 8KC. By 2000, frequency conversion technology using IGBT as power switching devices began to appear in China, with an operating frequency of up to 20KC and a power of up to 300KW. At present, 50KC and 100KW all-solid-state power supplies have appeared in China. It can be said that static frequency conversion technology is currently in a stage of rapid development in China.

Use and protection of thyristors

Thyristor is the abbreviation of crystal thyristor, which is a general term for various switching devices with a PNPN four-layer structure. According to the definition of the International Electrotechnical Commission (IEC), thyristors refer to semiconductor devices with more than 3 PN junctions, and the main voltage-current characteristics have two stable states of conduction and blocking in at least one quadrant, and can be converted between these two stable states. The thyristor we usually say is one of them, collectively known as Silicon Controlled Rectifier, mainly ordinary thyristor (KP), fast thyristor (KK), high-frequency thyristor (KG), bidirectional thyristor, gate turn-off thyristor (GTO-Gate Turn Off Thyristor) and light-controlled thyristor (LTT-Light Triggered Thristor).

Application of thyristor

Due to technical reasons, the voltage and current capacity of a single thyristor is limited and often cannot meet the requirements of high power. In order to solve this problem, two or more thyristors must be used in series or parallel operation. Due to the limitations of process conditions, the characteristic parameters of the thyristor itself are different. When the thyristors are connected in series or parallel, strict measures must be taken to limit the current and voltage differences within the allowable range to ensure the reliable operation of each thyristor.

Thyristor series connection

Usually two thyristors are used in series to solve the problem of insufficient withstand voltage of a single thyristor. This requires solving the problem of average distribution of the thyristor's working voltage, including static voltage balancing and dynamic voltage balancing. Static voltage balancing can be solved by using non-inductive resistors in series to divide the voltage; dynamic voltage balancing is more complicated because: the difference in component parameters dv/dt and the difference in reverse recovery time lead to uneven distribution of voltage on the components during the opening and closing process. In extreme cases, the branch voltage can be added to one thyristor.

This problem can be solved by connecting capacitors in parallel to limit dv/dt. However, in fact, during the component opening process, the capacitor discharges through the component to affect di/dt. Usually, a resistor is connected in series with the capacitor to form an RC resistor-capacitor absorption voltage-equalizing circuit. In order to limit the surge current on the branch, a saturated inductor or magnetic ring is usually connected in series on the branch, thus forming a circuit structure as shown in the figure.

Parallel connection of thyristors

Due to the improvement of the withstand voltage level of a single component, it is more common for each component to work in parallel to increase the power of the equipment. Take two thyristors working in parallel as an example, as shown in Figure 2.

Ideally, the current distribution is I1=I2=I/2. However, due to differences in component parameters, such as differences in saturation conduction voltage drop, differences in di/dt, and differences in distributed inductance caused by subtle differences in circuit installation processes, I1≠I2 is directly caused. In severe cases, components with large currents will burn out due to overcurrent. Therefore, measures must be taken to ensure that the difference between I1 and I2 is within the allowable range.

The commonly used methods are: 1) Use conjugate inductors to ensure dynamic current sharing; 2) Parallel RC circuits to absorb surge voltages; 3) Try to use thyristors with consistent on-state voltage drops to work in parallel; 4) Strictly follow the installation process to ensure that the distributed inductance of each branch is as consistent as possible, as shown in Figure 3.

The protection measures mentioned above should be treated differently according to the operating frequency of the components. In the three-phase rectifier circuit, a delta-shaped RC filter is usually added at the power supply end.

Due to the characteristic parameters of the thyristor itself, its maximum operating frequency is generally limited to below 8KC. For the use requirements of higher frequencies, super-audio power supplies that use IGBT as power switching elements have appeared in China.

Use and protection of IGBT

Insulated Gate Bipolar Transistor (IGBT or IGT—Insulated Gate Bipolar Transistor) is a new type of composite device developed in the mid-1980s. IGBT combines the advantages of MOSFET and GTR, so it has good characteristics. At present, the current/voltage level of IGBT has reached 1800A/1200V, the turn-off time has been shortened to 40ns, the operating frequency can reach 40kHz, the holding phenomenon has been improved, and the safe operating area (SOA) has been expanded. These superior properties make IGBT an ideal power device for power electronic devices such as high-power switching power supplies and inverters. The driving method of IGBT is significantly different from that of thyristor, resulting in a great difference in the control circuit. Thyristor is driven by a narrow pulse signal with a strong rising edge, while IGBT is driven by a square wave.

IGBT requirements for gate drive circuits

The static and dynamic characteristics of IGBT are closely related to the gate drive conditions. The gate positive bias voltage +VGE, negative bias voltage -VGE and gate resistance RG have different degrees of influence on the on-state voltage, switching time, switching loss, short-circuit withstand capability and dvce/dt parameters of IGBT.

The positive bias voltage +VGE provided by the gate drive circuit to the IGBT turns on the IGBT. In practical applications, +15V is usually used, considering the factors such as the turn-on time, turn-on loss, and the time the device withstands the short-circuit current when short-circuited. The negative bias voltage -VGE provided by the gate drive circuit to the IGBT turns it off. It directly affects the reliable operation of the IGBT. In order to prevent the IGBT from dynamically holding, the negative gate bias voltage should be -5V or lower. The magnitude of the negative bias voltage has little effect on the turn-off time loss.

In addition, the gate drive voltage must have a sufficiently fast rise and fall speed to turn the IGBT on and off as quickly as possible to reduce the turn-on and turn-off losses. After the device is turned on, the drive voltage and current should maintain a sufficient amplitude to ensure that the IGBT is in a saturated state. Since IGBTs are mostly used in high-voltage and high-current applications, the signal control circuit and the drive circuit should be isolated by a high-speed optoelectronic isolation device with strong anti-interference ability and short signal transmission time. In order to improve the anti-interference ability, the lead from the drive circuit to the IGBT module should be as short as possible, and the lead should be double-glue wire or shielded wire.

IGBT protection measures

Since IGBT has extremely high input impedance, it is easy to cause electrostatic breakdown. When IGBT is used for power conversion, in order to prevent abnormal phenomena from causing device damage, the following protection measures are usually adopted:
1) Cut off the gate signal through the detected overcurrent signal to achieve overcurrent protection;
2) Use the buffer circuit to suppress overvoltage and limit excessive dv/dt;
3) Use the temperature sensor to detect the shell temperature of the IGBT. When the allowable temperature is exceeded, the main circuit trips to achieve overheating protection. Since IGBT has a positive temperature coefficient and good parallel working characteristics, IGBT often uses multiple components to work in parallel. There are no other special requirements for the main circuit except symmetry.

Judging from the current usage, the static frequency conversion power supply that uses IGBT as the switching element has a significantly lower failure rate, fewer damaged components, and lower maintenance costs. It is a new development direction for static frequency conversion technology.