Free wheeling diode

[Pax Writer]

Hello users

I don't know if this is the right place to ask, but here goes:
I need to find a free wheeling diode to sit in parallel with a brushed DC-motor to deal with the reverse voltages occurring when the motor is stopped.
The motor will have a peak consumption of around 30-40 amps and run at 8-12V.
The diode should be SMD - Preferebly some SMD-equivalent to TO220 or smaller.
What kind of general characteristics must such a diode have, and can you suggest any actual types.
I stand by to be educated
Thanks in advance.

Pax

 

[wmmullaney]

Try googling

 

[mneary]

reverse voltages occurring when the motor is stopped

Do you mean reverse voltages when the circuit is trying to stop the motor? This probably needs an active circuit, since it needs to be disabled when you intend to drive the motor.

 

[saturn1bguy]

Do you mean reverse voltages when the circuit is trying to stop the motor? This probably needs an active circuit, since it needs to be disabled when you intend to drive the motor.

I think he means back EMF from the motor when the circuit turns off
(is that right, Pax Writer?). I don't know if an SMD diode is what you're
looking for, it sounds like you have some really beefy motors. You may
want a discrete device right at the motor...

You mentioned peak motor current, what's the normal operating current?

Corey

 

[saturn1bguy]

What kind of general characteristics must such a diode have?

As Mneary recently stated to someone else, "The diode must have a reverse
voltage rating of more than 12V, and a forward repetitive pulse current rating
that's more than your [nominal motor] current."

If I'm out of line for quoting, please forgive me

 

[dknguyen]

"reverse voltages occurring when the motor is stopped"
THat is called flyback (or the voltage that rises as the free wheel current tries to find a path to flow).

Parallel with the motor you say? So this is a unidirectional motor circuit then? If it's an H-bridge you can reduce the diode requirements by changing the order of the transistors when you switch the H-bridge on and off. BY complimentary switching (which only uses the diodes to pass the flyback current until some MOSFETs can turn onto pass the flyback current instead, and relieve the load on the diode) or alternating switching between the high side and low side transistor (which alternates the flyback current path between different diodes rather than forcing it all through a single diode).

You want a very low [forward] voltage diode so the diode the flyback voltage will exceed the battery voltage as little as possible before the diode can turn on, and as fast as possible so the flyback current can start flowing as quickly as possible once the diode is able to turn on to stop the flyback voltage from rising any farther. ie. schotky diodes.

 

[saturn1bguy]

You want a very low voltage diode...

Just to be sure, dknguyen means a diode with a low FORWARD voltage.
That's a separate issue from ensuring the freewheel diode sports a reverse
breakdown voltage that's higher than your power supply...just to be clear

Corey

 

[dknguyen]

Just to be sure, dknguyen means a diode with a low FORWARD voltage.
That's a separate issue from ensuring the freewheel diode sports a reverse
breakdown voltage that's higher than your power supply...just to be clear

Corey

Ah, thanks for catching that. Corrected now.

 

[Pax Writer]

Hey folks

Thanks for your answers.
The circuit in which the diode is needed can be found here:
https://www.electro-tech-online.com/threads/simple-pwm-motor-controller.33974/
Look at the illustration uploaded by Audioguru (fifth post from below). This shows that a MOSFET is inserted serially with the motor, which also has a diode in parallel to handle the back EMF (thanks for clarifying this term for me, by the way). The motor is very similar to the types used in RC model race cars. The peak consumption of ~40 amps is a cautious guestimate, and I think that the motor's average consumption when running is around 20 amps. I can't be more specific about the motor because the manufacturer never releases any data sheets for the motor, and I don't have any equipment which can handle these currents.
In my prototype circuit, I'm using a BYW29 diode in a TO-220 package, but I'd prefer a type which is not leaded

 

[dknguyen]

Oh, 5th schematic down. I was staring at the one in the opening post and couldn't figure out what was going on (if anything). That there is a unidirectional motor driver.

Yeah, you can't mess around with switching to get the MOSFETs to share the flyback current with the a smaller flyback diode because you only have one switch. A diode with a peak current of your motor's stall current (or just maximum expected current) is probably good. COntinuous current rating to match is obviously better and more conservative.

Try a PDS1040 from Diodes Inc. perhaps.

 

[ronsimpson]

Search for diode.
www.digikey.com
Type diode in the search window.
Go to “Discrete Semiconductor Products”
Select “Diodes, Rectifiers – Single”
On the far right mark “15V through 25V” click “APPLY FILTERS”
On the far left mark “20A through 60A” click “APPLY FILTERS”
Click on a part # .
Click on “Technical/Catalog Information”.
Click on “Datasheet(s)”.

I was looking at one of the 20A diodes (SMT). It can handle hundreds of amps for a short time. If the motor is just turned on/off, a smaller diode will work. If the FET is used for speed control (turned on/off 20,000 times/second) then a larger diode should be used.

 

[Pax Writer]

@dkngyen: Thanks for the tip. I'll probably have to find a non-Danish supplier if I choose this one, though. I couldn't find it on my usual supplier sites. By the way, the girl in your avatar is very beautiful. Is that your girlfriend?

@ronsimpson: Thanks for helping me out. My main problem is not that I don't know how to search, the main problem is that I don't know which characteristics I should look for in a good free wheeling diode. This, however, gives me a clue. The motor speed is controlled with a 555-circuit - The frequency is around 150Hz.

 

[dknguyen]

@dkngyen: By the way, the girl in your avatar is very beautiful. Is that your girlfriend?

No, it's a CD case.

 

[Pax Writer]

Anticlimax - Good choice of case, though.

 

[Hero999]

No, it's a CD case.

Good, because if you had answered yes then I would have had to inform the police.

 

[Styx]

Hey folks

Thanks for your answers.
The circuit in which the diode is needed can be found here:
https://www.electro-tech-online.com/threads/simple-pwm-motor-controller.33974/
Look at the illustration uploaded by Audioguru (fifth post from below). This shows that a MOSFET is inserted serially with the motor, which also has a diode in parallel to handle the back EMF (thanks for clarifying this term for me, by the way).

The parallel diode of the MOSFET (the intrinsic diode) does NOTHING for the circuit, it is there because MOSFET's always have an anti-parallel diode (it might not be connected for use mind and its more a zener then a diode). The diode that allows the current in the machine's coil to carry on flowing is teh one parallel to the machine

Replace the machine with an inductor (effectivly what it is) and think about what happens when the FET opens and current is flowing through that inductor AND there is no anti-parallel diode...

THe fundemental equation for an inductor is: V = Ldi/dt.
You have just opened the switch to try to stop the current, in the ideal case it will open instantly forcing di/dt to tend to infinity, via this equation this produces an infinite voltage across the inductor

In practice the switch has a finite switching time to impose a finite di/dt thus producing a finite V. With no anti-parallel diode this V will just increase until it reaches the breakdown voltage of the FET.

With a diode in anti-parallel, when this induced voltage exceeds the voltage the diode needs to conduct it produces a zero-volt loop (well a 0.6V loop) for the current in the inductor to keep flowing


I personally don't like the term "BackEMF" when talking about open-circuiting a charged inductor (less so in cases like this involving electrical machines).
BackEMF is something that an electrical machine will generate when its rotor is rotating, I personally refer to the voltage in this case as the "induced voltage"


so just to claify the anti-parallel diode of the FET does nothing for this circuit (it would still work if it was removed). It is the anti-parallel diode of the motor that is crucial here

 

[saturn1bguy]

The parallel diode of the MOSFET (the intrinsic diode) does NOTHING for the circuit. The diode that allows the current in the machine's coil to carry on flowing is teh one parallel to the machine

Right.

That's the diode we're all talking about, the one parallel to the motor. No one has ever spoken to the instrinic diode. I think we all understand why the freewheel diode is required, too. Perhaps you looked at the wrong schematic. Here's the one we've all been referring to:

https://www.electro-tech-online.com/attachments/555-motor-pwm-speed-control-png.15607/

 

[Styx]

well then Pax Writer didn't word it nicely so my parcer could cope

This shows that a MOSFET is inserted serially with the motor, which also has a diode in parallel to handle the back EMF

I followed that as: MOSFET in circuit which also as a diode in parallel to handle the backEMF. it wasn't clearly stated that the part being refered to had been changed

 

[saturn1bguy]

well then Pax Writer didn't word it nicely so my parcer could cope

True.

But a glance at the schematic in question would have resolved this for you, as well as the fact that he's asking to ADD a diode to the circuit, and not asking for clarification on the presence of the FET's intrinsic diode.

FYI,
Corey

 

[Pax Writer]

Hello Styx and Saturn, I'm sorry if I phrased myself poorly, but thanks for your help any way. The problem of finding a suitable diode has been solved now, but there's always something interesting to be learned from the comments of my wiser colleagues
Next time I'll be sure to think twice about the wording before I click the magic button.
Cheers, Pax.