Can the freewheeling diode be omitted when the MOS tube controls the solenoid valve?

sfcsdc

Can the freewheeling diode be omitted when the MOS tube controls the solenoid valve? [Copy link]

 

As shown in the figure, the voltage at the source of the diode will be equal to 24V + the voltage generated by the inductor of the solenoid valve when the solenoid valve is reset.

If this voltage can be ensured to be within the withstand voltage range of the MOS tube, can the freewheeling diode be omitted?

Are there any risks in doing this?

maychang

When the current in the solenoid valve coil is suddenly cut off, the induced voltage generated can be ten times or even more than the power supply voltage.

If the MOS tube selected by the OP can withstand the voltage, the freewheeling diode can be omitted. However, such a MOS tube is much more expensive than a MOS tube with a lower voltage resistance, and is more expensive than a MOS tube with a lower voltage resistance plus a diode. In addition, the on-resistance of a high-voltage MOS tube is also large, which will increase the loss of the MOS when it is turned on and generate more heat.

Using a MOS tube with a lower withstand voltage and a freewheeling diode is the best choice in terms of cost and is also safe.

maychang

In addition, it is not wise to use 1N4148 as the freewheeling diode. Ordinary rectifier diodes such as 1N4001~4007 should be used.

qwqwqw2088

If it is to control the solenoid valve, this circuit is wrong. The diode is not connected in series with the MOS tube, but should be connected in parallel at both ends of the solenoid valve. 1n4007 will be enough.

qwqwqw2088

The current is not large, so you don't need it.

MOS tube controls relays, solenoid valves, motors and other coil-type inductor devices

DC control, the reverse electromotive force generated by the coil power failure can be prevented from damaging the drive circuit

PowerAnts

maychang posted on 2020-4-8 10:02 In addition, it is unwise to use 1N4148 as the freewheeling diode. Ordinary rectifier diodes such as 1N4001~4007 should be used.

If the requirements for the shutdown response are not high, I have an idea to save money for the OP. What do you think, Mr. Zhang?

Increase the gate RC constant to control the slope of the current drop. According to e=L*di/dt, assuming the relay inductance is 10H, the coil current is 50mA, and the transition time is 0.1 seconds, the induced voltage is only 5V.

sfcsdc

maychang posted on 2020-4-8 10:00 When the current in the solenoid valve coil is suddenly cut off, the induced voltage generated can be more than ten times the power supply voltage or even more. If the MOS tube selected by the host can withstand voltage...

Thank you, but the inductance of the solenoid valve currently used is not large. Actual testing with an oscilloscope shows that during normal operation time, the voltage will only rise by about 7V.

How can it reach more than ten times the power supply voltage?

chunyang

sfcsdc posted on 2020-4-8 10:59 Thank you, but the inductance of the solenoid valve currently used is not large. Using an oscilloscope to test it, the voltage will only rise by about 7V during normal operation time. How...

The magnitude of the voltage jump that can be observed is also related to the bandwidth of the oscilloscope.

zxy2266

This picture seriously misleads you. The motor will generate a large induction current during use, so it is best to add it. If you are not afraid of burning, 4148 should be replaced with a high current and high voltage tube, and a high frequency one is better.

chunyang

I suggest the OP to follow the conventional approach and save even the money for a rectifier diode. I am not willing to buy such a product and even think it should be boycotted.

maychang

PowerAnts posted on 2020-4-8 10:51 If the requirements for the turn-off response are not high, I have an idea to save money for the OP. What do you think? Make the gate RC constant larger, ...

You mean, let the coil inductance discharge slowly.

Then we also need to see which one has a higher cost, adding RC or adding a diode.

maychang

sfcsdc posted on 2020-4-8 10:59 Thank you, but the inductance of the solenoid valve currently used is not large. Using an oscilloscope to test it, the voltage will only rise by about 7V during normal operation time. How...

"During normal operation, the voltage will only rise by about 7V"

I said ten times on the second floor, but it is just an estimate. The actual voltage spike depends mainly on the distributed capacitance of the winding.

Your oscilloscope will only observe a 7V rise. This value is also related to the bandwidth of your oscilloscope. If the bandwidth is relatively low, the observed voltage spike will be relatively small.

maychang

sfcsdc posted on 2020-4-8 10:59 Thank you, but the inductance of the solenoid valve currently used is not large. Using an oscilloscope to test it, the voltage will only rise by about 7V during normal operation time. How...

Diodes like 1N4007 cost only a few cents, so there is really no need to save them. It should be noted that saving diodes may damage the MOS tube, which is not worth it.

threetigher

chunyang posted on 2020-4-8 11:05 I suggest that the OP should follow the conventional way and save even the money of a rectifier diode. I don't want to buy such products and even think they should be boycotted. ...

Correct answer!

PowerAnts

There is another situation: the diode is only used here on the board, and this material with a usage of 1 is very annoying. Then, capacitors are always used on PCB!

The measured resistance of a 24V relay coil is 480 ohms, the inductance is 330mH at 100HZ, 180mH at 1KHz, and the middle value is 250mH. Critical damping is required to avoid overshoot at all, and the loop characteristic impedance should be 240 ohms. A 4.35uF capacitor can be connected to the relay or MOSFET. If a little overshoot is acceptable, 1uF is enough, which is commonly used on PCB.

maychang

PowerAnts posted on 2020-4-8 11:21 There is another situation: the diode is only used here on the board. This material with a usage of 1 is very annoying. Then, capacitors will always be used on PCB! ...

The capacitor placed at the drain end of the MOS tube has a much larger capacity than the capacitor placed at the gate end of the MOS tube. If you want to save a little, it is better to place the capacitor at the gate end of the MOS tube.

PowerAnts

maychang posted on 2020-4-8 11:09 You mean, let the coil inductance discharge slowly. Then we have to see which one is more expensive, adding RC or adding a diode.

The RC is already in the original poster's circuit diagram.

PowerAnts

maychang posted on 2020-4-8 11:24 Placing the capacitor at the drain end of the MOS tube has a much larger capacity than placing the capacitor at the gate end of the MOS tube. If you want to save a little, it is better to place the capacitor at the gate end of the MOS tube. ...

Makes sense!

maychang

PowerAnts posted on 2020-4-8 11:31 The OP already has RC on his circuit diagram

Yes, there is already RC in the picture in the first post of the OP.

The OP observed that "the voltage only rises by about 7V" on the oscilloscope, which is probably the effect of this RC. If this is indeed the case, then there is no need to add a freewheeling diode, as long as this RC is reliable and not damaged.

ddllxxrr

It cannot be removed. It will burn the MOS tube.