[SOLVED] Select the right Freewheeling diode.

[yokohama]

Hi everybody,

How can I choose the correct freewheeling diode in the circuit attached.
V power = 24V
Resistance for the solenoid: 4 Ohms.



Thank's.

 

[Nasser750gx]

I would go for a fast Schottkey.

 

[Tunelabguy]

I agree about the fast Schottkey. You can also say that the diode must be able to handle 6 amps intermittently, and withstand a reverse voltage of at least 24 volts.

But I want to comment on your "CS" output from the source of the power FET. I assume that stands for "current sense". Since you are driving this circuit with a PWM signal and the solenoid is inductive, the current through the solenoid is going to be a somewhat constant current that rises slightly when the FET is on and falls about the same amount when the FET is off. But it does not fall to zero, because the freewheeling diode allows the current to continue to flow during the time when the FET is off. But your CS output only reflects the current when the FET is on. The CS output fall abruptly to zero when the FET is off. Therefore the CS output is not a very good representation of the current in the solenoid. In fact, if you filter the CS output, it will represent the true solenoid current multiplied by the PWM duty cycle fraction. As the duty cycle changes, the CS output will change, and much more than the average solenoid current changes.

When I had to do something like this, I put my 0.1 Ohm current sense resistor directly in the solenoid connection. That means the voltage signal from the current sense resistor is now differential, instead of single ended and ground referenced. Therefore I had to use a somewhat exotic opamp circuit with very good CMRR and common mode range that included the supply rail. Not easy, but not too hard either. And it does give me the true current through the solenoid all the time - both when the FET is on and when it is off.

 

[yokohama]

@Tunelabguy:
Do you mean that I must put the 0.1 Ohm current sense resistor close to the solenoid casing ? If this is what you mean, it's not possible in my case because the solenoid can be changed, and can have a different resistance (greater than 4 ohms).

When I had to do something like this, I put my 0.1 Ohm current sense resistor directly in the solenoid connection. That means the voltage signal from the current sense resistor is now differential, instead of single ended and ground referenced. Therefore I had to use a somewhat exotic opamp circuit with very good CMRR and common mode range that included the supply rail. Not easy, but not too hard either. And it does give me the true current through the solenoid all the time - both when the FET is on and when it is off.

I'm not sure that I've well understand this section can you please be more precise or can you suggest a diagram ou schematic ? Thank's. Because I think It's very important.

 

[Tunelabguy]

I mean the current sense resistor must be in place of the wire that goes from the anode of the diode to the bottom of the solenoid.

 

[yokohama]

you mean like this ?

 

[crutschow]

Any diode of sufficient voltage and current rating will work since all standard junction diodes have a fast turn-on time, which is what you want.
A fast turn-off and lower forward drop, which a Schottky provides, is not required and has no advantage when inductive transients are suppressed by placing a diode across the inductance.


Whether you need to move the sense resistor or not depends upon what you are trying to sense.
If it's just to limit the peak solenoid current then your original circuit is fine.

 

[FvM]

Regarding current sense options. I presume Tunelabguy meaned to place the shunt at the +24V side of solenoid, which avoids large common mode AC and swing above supply rail.

We should ask however about the purpose of the current sense. Sensing the total solenoid current may be necessary if you want to measure the current exactly. The low side shunt measurement can be still mapped to total solenoid current by considering the duty cycle setting. If it's e.g. intended as an overcurrent protection for the MOSFET, the low-side current can be used without further calculation.

Regarding diode rating, consider that the average diode current can't be larger than 1.5 A, a 3 A diode e.g. MBRS340 is sufficient.

 

[yokohama]

@ FvM and crutschow

The purpose of the shunt resistor is overcurrent protection. I measure the current (voltage between GND and CS) only when the PWM signal level is high. This way I have a precise measure of the current across the solenoid and the mosfet.
What do you think about this approach of current measurement.
MBRS340 is ok for me and I think RK44 is the equivalent on through hole version.

 

[FvM]

Sounds reasonable.

 

[yokohama]

Thank's again FvM.

 

[Tunelabguy]

I agree with FvM that for overcurrent protection, your existing low-side shunt is sufficient, and a lot simpler than high-side current measurement, like I had to do in a similar project, which was a solenoid test stand.