Reverse protection diode

[anishkgt]

Hi All,

I have a power supply that powers a servo and the forward voltage is 7v and takes about 1-2A. Would this diode do the job https://www.digikey.com/product-detail/en/stmicroelectronics/BAT60JFILM/497-3707-1-ND/686425

[Ian.M]

Not if its in series.  Look at the Average forward current versus ambient
temperature graph in its data sheet. Its only good for about 0.5A @ 40 deg C.

[anishkgt]

Thank you Ian for pointing it out to me.

I was just filtering from the list with what i knew with working servo. I know it works on 7v and consumes approx 1A. What more should i look for when selecting a reverse polarity diode ?

[NolanM]

Have you considered using a MOSFET to do reverse-polarity protection?

Here's a good explanation:

https://electronics.stackexchange.com/questions/61692/reverse-polarity-protection

[bjcuizon]

Here is mr.electroboom's recommendation of a reverse polarity protection circuit:
Watch this.

[anishkgt]

MOSFET and transistor are another way to accomplish it but the diode is much more simpler and cost effective too.

What is "voltage - Forward (Vf) (Max) @ if" when selecting a reverse polarity. Something that i am lost with.

[bjcuizon]

MOSFET and transistor are another way to accomplish it but the diode is much more simpler and cost effective too.

But sometimes it will just blow the fuse and  not protect the circuit at all and cause permanent damage.
However, I agree with you if your cost cutting. The diode is simpler and cost effective.

[anishkgt]

Well blowing the fuse or burn a trace [emoji16] which happened when I had. Want to prevent that in the future so was hoping if somebody could suggest one for me.


Sent from my iPhone using Tapatalk

[Ian.M]

There are three common ways of implementing reverse polarity protection:

• Series Diode
• Shunt Diode + Fuse or other over-current trip
• Series MOSFET

Lets discuss (1) a series diode:

What is "voltage - Forward (Vf) (Max) @ if" when selecting a reverse polarity. Something that i am lost with.

It all depends on whether your load can tolerate the Vf forward voltage drop at the expected load current.  If it can't, and you've already found the lowest Vf highest If diode that you can afford or that will fit, then you have to choose a different method.   The datasheet will give some worst case 'spot' values of Vf for various If, and better manufacturers datasheets also include a typical Vf vs If graph.  However that isn't a worst case graph so you may have to plot whatever worst case figures you have on the graph, then scale it to get an intermediate value for you actual peak load current then allow a generous safety factor because you are extrapolating from the original limiting data.  Obviously the diode also dissipates power so you need to work out whether the package can dissipate Vf*If at your expected If and maximum ambient temperature.

For (2) the shunt diode, everything is *FAR* more scary!    The diode is no longer in series with the load so its Vf drop no longer subtracts from the supply voltage.   However the diode must now survive an extremely high pule If, big enough to blow the fuse, which is probably well outside the specs given in the datasheet.  If you are making a few million units, contact the diode manufacturer for a recommendation for a diode that can survive a non-repetitive surge of hundreds of Amps to blow a fuse of a specified characteristic but is still in a small enough package, but the rest of us have to do our own destructive testing to establish suitability.   If you pick a diode thats too small, it will probably fail shorted when it blows the fuse, and continue to blow the fuse even when the reverse polarity is corrected.    If you pick one that's far too small it will short, then blow open before the fuse blows and your product will DIE!, probably with lots of smoke and maybe even flames 

A further complication of (2) is that your device 'sees' the Vf of the diode at the peak surge current as reverse voltage.  You need enough die area to keep this voltage low enough to avoid damage.

[TheUnnamedNewbie]

I just answered this exact question on stackexchange. I'll just link it in here instead of copy-pasting, in case others provide better answers still.

https://electronics.stackexchange.com/questions/326192/a-question-about-preventing-reverse-polarity-for-supply-rails-of-an-opamp

[Yansi]

Have you considered using a MOSFET to do reverse-polarity protection?

Here's a good explanation:

https://electronics.stackexchange.com/questions/61692/reverse-polarity-protection

Something ain't right with that circuit. Maybe using that kind of symbol for MOS-FET transistor in the first place.

[TheUnnamedNewbie]

Something ain't right with that circuit. Maybe using that kind of symbol for MOS-FET transistor in the first place.

This matter has been coverd again and again. IC designers use this symbol often, whether you like it or not. I also prefer the "correct" symbol but when doing a quick schetch, those symbols are nice since they are easier and faster to draw, and less complex when you are drawing small schematics.

[Yansi]

Understand that, but supposed the arrow goes in like that, seems they have drawn the wrong polarity of the transistor, didn't they? I'd expect the arrow should follow the substrate diode.

And according to this, they have indeed drawn it wrong!


Just look at the original schematic:


It will not work with NMOS like that.

[anishkgt]

So i checked with customer support at digikey and i was able to filter down to this https://www.digikey.com/products/en/discrete-semiconductor-products/diodes-rectifiers-single/280?k=diode&k=&pkeyword=diode&pv915=48&pv913=354&FV=1c0002%2C1800018%2Cmu10V%7C2071%2Cmu12V%7C2071%2Cmu8V%7C2071%2Cffe00118%2Ce480013&mnonly=0&ColumnSort=0&page=1&stock=1&quantity=0&ptm=0&fid=0&pageSize=25

If am correct would the attached be the correct way to connect the diode. Pin 5 is the GND of the power supply and PIN 4 is +7v. Pins 1 2 3 are PWM, VCC and GND respectively.

[TheUnnamedNewbie]

Understand that, but supposed the arrow goes in like that, seems they have drawn the wrong polarity of the transistor, didn't they? I'd expect the arrow should follow the substrate diode.

And according to this, they have indeed drawn it wrong!

I don't want to say that image is wrong as to my knowledge there is no standard on these symbols, but it is reversed to what I am used to seeing. The arrow doesn't follow the substrate diode. This is what I'm used to seeing as on wikipedia: https://en.wikipedia.org/wiki/MOSFET#Circuit_symbols:



If I am not mistaken, these symbols were originally used by Sedra and Smith. They used them in such a way that they were similar to the "same type" of BJTs, so the arrow is the source terminal, pointing in the same direction as the emitter of a BJT of the same type. So in the specific example, the device is a PMOS with the body diode correctly biased.

I have never come accross the reverse symbol on things like IEEE Xplore, so I would argue that the choice made in your example is the "less common" one.

That said, I encourage people to use the full symbol, and if that isn't possible, to use the 'digital' symbol (no arrow, and a "inverter" circle to represent pmos) if it's a quick sketch. I don't like the way the Sedra & Smith symbol implies there is a difference between source and drain (especially when the bulk is not connected to either). I have already seen many students confused by this.