Problems with two types of channel MOS tube circuits

shaorc

As shown in the figure below, NMOS and PMOS are used as a protection circuit to prevent reverse connection of the power supply. [1] The circuit diagram of the PMOS tube is easy to understand. For the circuit diagram of the NMOS tube (the top figure), the input Vin is directly connected to its gate, but the Vgs of the MOS tube is limited. If the input voltage Vin is too large, can the NMOS tube not be used for reverse connection protection? The PMOS tube does not have such a problem, but the on-resistance Rds of the PMOS tube is greater than that of the NMOS tube. [2] In the two-channel MOS tube reverse connection protection circuit, resistors are added to the gate. What is the role of it? [3] Since the two-channel MOS tube reverse connection protection circuits both use the conduction effect of the parasitic diode of the MOS tube itself, then in the push-pull output topology at the bottom of the figure, why is the role of the parasitic diode not considered? In the H full-bridge topology, is the diode connected in parallel with the MOS tube only regarded as a freewheeling function?

maychang

The input Vin is directly connected to its gate, but the Vgs of the MOS tube is limited. If the input voltage Vin is too large, can we not use the NMOS tube for reverse connection protection? Yes. If Vin is too large, corresponding measures need to be taken.

maychang

"PMOS tubes do not have such problems, but the on-state resistance Rds of PMOS tubes is greater than that of NMOS tubes." P-channel MOS tubes also have the problem that special measures must be taken to protect the MOS tube gate when Vin is large.

maychang

It is better to use Schottky diodes to replace MOS tubes for these two MOS tube anti-reverse connection circuits. It is simple and reliable and can adapt to a wider range of power supply voltages.

maychang

"In the push-pull output topology at the bottom of the figure, the effect of the parasitic diode is not considered?" In the push-pull output topology at the bottom, the N-channel tube seems to be drawn incorrectly.

shaorc

maychang published on 2019-5-11 07:02 Instead of these two MOS tube anti-reverse connection circuits, it is better to use Schottky diodes to replace MOS tubes. They are simple, reliable, and adaptable to a wider range of power supply voltages.

What is the forward conduction voltage drop of a Schottky diode? I have checked some online information. Although it is lower than that of ordinary diodes, it is also 0.4v. In this way, if the current is large, the power loss of the diode is not ideal. So I want to use nmos with a smaller on-resistance across the board. But as mentioned before, if the input voltage is too large, the gate must be protected. It seems that each has its pros and cons. If my input is only low voltage, such as a maximum of 24 volts and a minimum of 12 volts, then there are more choices, but I am confused, because the reverse voltage of a general Schottky diode is 60v, which can also meet the requirements.

shaorc

maychang published on 2019-5-10 22:35 The input Vin is directly connected to its gate, but the Vgs of the MOS tube is limited. If the input voltage Vin is too large, can we not use NMOS tube for feedback?

The measure to protect the mos gate can be resistor voltage division to achieve the purpose of voltage reduction?

shaorc

maychang published on 2019-5-11 07:04 "In the push-pull output topology at the bottom of the figure, the role of the parasitic diode is not considered?" In the push-pull output topology at the bottom, the N-channel tube seems...

Yes, the n-tube below me should have the diode direction reversed, so there is a question, if the positions of the power supply and ground are swapped in the push-pull output in the third figure of the first post, will the power supply be directly connected to the ground because the directions of the two parasitic diodes are in the same direction?

maychang

shaorc posted on 2019-5-11 09:15 Can the measure to protect the MOS gate be resistor voltage division to achieve the purpose of voltage reduction?

"Can the measure to protect the MOS gate be resistor voltage division to achieve the purpose of voltage reduction?" Generally, resistor voltage division can be used. The purpose of voltage division is to ensure that the voltage between the gate and source of the MOS tube does not exceed the allowable value, and at the same time the MOS tube can be fully turned on.

maychang

shaorc posted on 2019-5-11 09:21 Yes, the n-tube below me should have the diode direction reversed, so there is a question, if the power supply and ground are swapped in the push-pull output of the third picture in the first post...

"If the power supply and ground are swapped in the push-pull output of the third picture in the first post, will the power supply be directly connected to the ground because the directions of the two parasitic diodes are in the same direction?" You will know if you draw the diode inside the MOS tube correctly.

shaorc

maychang published on 2019-5-11 09:28 "If the positions of the power supply and ground are swapped in the push-pull output of the third figure in the first post, will the two parasitic diodes be in the same direction? ...

Yes, I understand it since I drew the correct direction. If the rated supply voltage of the chip or circuit is 24V, it must be able to withstand an overvoltage shock of 36V input for one hour and an overvoltage shock of 48V input for one minute. I think it is better to design it to be higher than 24V input and directly disconnect the input through the protection circuit. How are the 36V for one hour and 48V for one minute reflected? Because when it exceeds 24V, I need to use a protection circuit to disconnect the power supply. And for the overvoltage protection circuit, is it better to use power electronic devices such as diodes, voltage regulators, and transistors, or to use an overvoltage protection integrated chip + peripheral circuit topology?

maychang

shaorc posted on 2019-5-13 11:07 Yes, since I drew the correct direction, I understand that if the rated supply voltage of the chip or circuit is 24V, it must be able to withstand 36V input...

"How is this 36V for one hour and 48V for one minute reflected?" I have never seen such a regulation. For the test of insulation performance, there is a regulation of industrial frequency AC for one minute.

maychang

shaorc posted on 2019-5-13 11:07 Yes, since I drew the correct direction, I understand that if the rated supply voltage of the chip or circuit is 24V, it must be able to withstand 36V input...

And for the overvoltage protection circuit, is it better to use power electronic devices such as diodes, voltage regulators, and triodes, or to use an integrated chip with overvoltage protection + peripheral circuit topology? I think it is better to use discrete components, because discrete components can withstand overvoltage better than integrated chips.

bigbat

I have verified this circuit. It is usable. It is recommended that if the R3 resistor is not added, there will be a reverse voltage of about 2V when reversely connected. I don't know why. R1 and R3 in the figure are a voltage divider circuit. R1 is about 1K, R3 100K. It works well now.

shaorc

bigbat posted on 2019-5-26 10:41 I have verified this circuit. It can be used. It is recommended that if the R3 resistor is not added, there will be a reverse voltage of about 2V when reversely connected. I don’t know...

It should be R2 and R3 that form a voltage divider circuit. Also, if R3 is not added, there will be a reverse voltage of nearly 2V when reversely connected. Here, R1 should not be added, and R3 is indispensable.

zhuyebb

"In the push-pull output topology at the bottom of the figure, the effect of the parasitic diode is not considered?"