MOS drive circuit and bootstrap capacitor analysis

MOS driving circuit and bootstrap capacitor analysis mind map, including methods of driving MOS: fast charging, fast discharging, and its driving circuit.

Edited at 2021-12-21 17:56:10

PlotWizard

Recent works View more works>>

MOS drive circuit and bootstrap capacitor analysis

PlotWizard

Recent works View more works>>

Recommended to you
Outline

29. MOS drive circuit and bootstrap capacitor analysis

How to drive MOS

It is not only the main circuit that generates dv/dt and di/dt, but also the driving circuit. According to the characteristics of the MOS tube device, we know that to turn it on and off, it is necessary to quickly charge and discharge the Cgs capacitor.

fast charging

Depending on the source, loop, and impedance, you must first have a source, usually 12V or 15V. Leave a little more margin, because in the final circuit, there will be a little voltage drop from the 12V or 15V source to Cgs. The voltage on the Cgs capacitor cannot rise too fast, otherwise, oscillation will occur in the Miller plateau area in the actual circuit. Therefore, there must be a resistor, but this resistor cannot be too large.

The selection criteria for R1: actual measurement is required, and the absence of oscillation is the standard. Here is an empirical value: dozens of Ω~330Ω

Add a switch here to control its on and off, but this is only the conduction condition of the MOS tube drive circuit. If you want to turn off the MOS tube quickly, you must let Cgs discharge quickly.

rapid discharge

The fastest way is to directly connect a switch to GND, because when Q1 is turned off and Q2 is turned on, Cgs will act as a source and return to GND through R1 and Q2.

Drive circuit

When we connect the bases of Q1 and Q2 together, we can use a signal to control it. The circuit form formed by Q1 and Q2 is called a push-pull circuit, which is just a name. Its essence is still a triode. The MOS tube is a voltage-controlled device, but when it is turned on, it still needs a certain current driving capability. Just like we say that the triode is a flow-controlled flow device. To turn on the triode, Vbe ≥ 0.7V

When the input PWM signal is high level, Q1 is turned on and Q2 is turned off. Cgs is in the charging state at this time, which is the opening process of the MOS tube. As a voltage-controlled device, MOS tubes do not require drive current after they are turned on. During the turn-on process, they also require a drive current of up to hundreds of mA to meet the requirements for fast turn-on.

Similarly, when the input PWM is low level, the Cgs capacitor will be discharged, and the discharge current can be as high as hundreds of mA, thus realizing the switching of the MOS tube.

An extra R2 is added, with a general resistance of 15~20K. The function of this resistor is to discharge the Cgs capacitor. Without this resistor, when the entire system is not powered, Q1 and Q2 will be in a disconnected state, then , static electricity and external interference will charge Cgs. If the Cgs voltage is charged to the conduction threshold voltage, the channel of the MOS tube will be established, and the MOS tube will be turned on at this time. Then, as soon as the power is turned on, the main circuit will be turned on, which will cause a safety accident. Therefore, We need to add a R2 discharge resistor. This resistor cannot be too small, otherwise the partial voltage formed by R1 and R2 will drop and the Cgs voltage cannot be charged. It does not need to be too large. The general resistance is 15~20K

Bootstrap circuit

For the MOS tube in the BUCK circuit, its S pole is not connected to GND. Not only is it not connected to GND, but it also generates dv/dt, because when the MOS tube is turned on, the potential at point A is 310V. 15V starts from the positive terminal and charges the Cgs capacitor through R1. However, when the MOS tube is turned on, the potential at point A becomes 310V. Since the potential can only go from high to low, it is impossible to charge the Cgs capacitor with 15V. It can't be charged anymore, so how can we maintain the Cgs voltage and keep the MOS switch on? According to the characteristics of MOS tube devices, we only need to look at the pressure difference of Cgs to turn on the MOS tube. What we look at is the pressure difference, which is a relative concept.

Adding a capacitor here is like a bucket. We know that the voltage of the capacitor cannot change suddenly. In fact, more strictly speaking, the voltage difference cannot change suddenly. When the MOS tube is turned on, dv/dt will be generated at the S terminal (point A). It does not matter if the voltage suddenly changes. Since the voltage across the capacitor cannot change suddenly, point B will also become 15V 310V at that moment. =325V. In this case, as the source between the MOS tubes GS, it still maintains 15V. Point B = 325V, point A = 310V, and the Vba voltage difference is still 15V. In this case, the potential mutation of point A caused by the conduction of the MOS tube , does not affect the normal conduction of the MOS tube

This capacitor does not have 15V power at the beginning. How to get it? Before the BUCK circuit does not work, the entire circuit has only one source: 310V. Therefore, the initial voltage of this capacitor can only be provided by 310V. In fact, it is indeed 310V that charges this capacitor.

310V provides output energy to the bootstrap capacitor through R3 and D2. However, it only provides initial charging energy. In the future, it is not needed to provide it, because after the BUCK power supply is established, the BUCK output can be used to provide The capacitor replenishes energy. R3 and D2 only provide initial energy to the capacitor. After the power supply is established, a steady stream of energy is provided through the output end of the BUCK circuit. Therefore, the charging current of this circuit is generally not too large, 1~2mA is sufficient.

After the BUCK power supply is established, the output end can provide a steady stream of energy, but a D3 must be added, because point B will have dv/dt following point A, so don't affect the output voltage.

Add a circuit to limit it so that there is no consumption until the initial charge reaches 16V Add a switch. Before the C1 capacitor is charged to 16V, there is no consumption. If it is continuously charged, this time is about ms level. When the C1 capacitor is charged to 16V, we turn on the switch to provide limited energy to the previous PWM drive circuit. The consumption of the PWM circuit is generally on the order of 10mA. We must ensure that the output voltage of BUCK is before the C1 voltage drops from 16V to 10V. If it can be established, that’s it.

In fact, the UC3842 chip is turned on at 16V and turned off at 10V. You only need to understand it qualitatively. You only need to add a voltage feedback loop, and then the theme framework of the entire BUCK circuit is set up. In terms of circuit function, it can output 15V normally, and the chip also has overvoltage protection, current limiting protection, overtemperature protection, etc.

For the BUCK circuit, if a voltage feedback loop is implemented, an optocoupler needs to be added to achieve isolation.

Summarize

What is the purpose of explaining BUCK power supply? The purpose is to let everyone know that there are dv/dt and di/dt interference sources in the main loop and drive loop.