Simple reverse polarity protection?

[shapirus]

The relay could do with a back-EMF diode.

that is definitely implied . it was only a quick and dirty example.

[Wolfram]

What's this got to do with relays? Perhaps you replied to the wrong thread.

If an ideal diode circuit is required, then how about a MOSFET plus a couple of BJTs?

The stackexchange thread linked below recommends the DMMT5401, which isn't expensive.
https://electronics.stackexchange.com/questions/223935/understanding-an-ideal-diode-made-from-a-p-channel-mosfet-and-pnp-transistors

All that's needed are some zener diodes to protect the MOSFETs' gates.

Correct me if I'm wrong, but wouldn't this circuit have issues with exceeding the reverse BE breakdown of Q1 when the input voltage is negative? The Q1 BE junction will be reverse biased in series with the forward-biased BE junction of Q2, and once this voltage exceeds some five-six volts you'd be outside of the maximum rating of most common small signal transistors. For the DMMT5401 specifically, max Vbe reverse voltage is 5 V.

[Vovk_Z]

I have a bad experience with a relay as a reverse protection. Relay contacts don't like to work with a DC current, so I guess a Mosfet-based design may work more reliably despite all Mosfet volnerabilities.

[Zero999]

What's this got to do with relays? Perhaps you replied to the wrong thread.

If an ideal diode circuit is required, then how about a MOSFET plus a couple of BJTs?

The stackexchange thread linked below recommends the DMMT5401, which isn't expensive.
https://electronics.stackexchange.com/questions/223935/understanding-an-ideal-diode-made-from-a-p-channel-mosfet-and-pnp-transistors

All that's needed are some zener diodes to protect the MOSFETs' gates.

Correct me if I'm wrong, but wouldn't this circuit have issues with exceeding the reverse BE breakdown of Q1 when the input voltage is negative? The Q1 BE junction will be reverse biased in series with the forward-biased BE junction of Q2, and once this voltage exceeds some five-six volts you'd be outside of the maximum rating of most common small signal transistors. For the DMMT5401 specifically, max Vbe reverse voltage is 5 V.

You're right. Well-spotted. I should know better than to copy circuits without thinking through how they can fail.  It's obviously not shown up in my simulation because V2 is no more than 5V away from V1. The obvious solution is to use another matched pair of transistors as diodes, to block the reverse voltage.

[BillyO]

Isn't the word simple in the title? 

[shapirus]

Isn't the word simple in the title? 

post #6 should've been the final one.

(ideal diodes is actually an interesting topic to discuss, but it doesn't really belong here.)

[Zero999]

Isn't the word simple in the title? 

I know, but I had to correct the error.

post #6 should've been the final one.

(ideal diodes is actually an interesting topic to discuss, but it doesn't really belong here.)

I disagree about it not belonging here. One of the good things about this place is threads drifting off-topic. You don't have to keep following a thread, if it no longer interests you.

[Xena E]

I have a bad experience with a relay as a reverse protection. Relay contacts don't like to work with a DC current, so I guess a Mosfet-based design may work more reliably despite all Mosfet volnerabilities.

That raises a very good point.
Switching at load can cause contact welding and also problems with sustained arcing on DC, designers must pay attention to manufacturers specification of the components... a contactor rated for swtching 240VAC at 20A for instance, isn't ever suitable for the same voltage and power at DC. Selection of fuses also follow this caution.

[BrianHG]

Just stumbled upon this LM74610 IC (https://www.ti.com/lit/ds/symlink/lm74610-q1.pdf) and it kind of seems like an amazing and simple reverse polarity protection IC with a somewhat high voltage support of 42v, but I'd love to find one that supports 60v? Anyone used it before (or similar) and recommend it? AFAIK, I don't know it would be possible to add external circuitry to this IC to allow for higher voltage support?

How about these guys with 65v automotive support (Back to back mosfet controller for polarity protection with ORing selection for multiple sources):
https://www.ti.com/product/LM7481
https://www.ti.com/product/LM7480
https://www.ti.com/product/LM74900-Q1

Note that 2 channel ORing controllers also exist for either multiple battery sources, or mains supply VS battery source with a forced switch selection input pin.
https://www.ti.com/power-management/power-switches/ideal-diodes-oring-controllers/products.html#p238max=65;100&p339=External%20single%20FET&p1498=Catalog&p1694=ORing%20controller

[BrianHG]

Don't forget about FERD diodes from STMicroelectronics.  You might find some in the 0.2v to 0.3v drop at reasonable currents and they cost less than the ideal diode controllers.  Though, the reverse leakage current can be as much as 1ma.

[Smokey]

https://www.ti.com/lit/ds/symlink/lm74502.pdf

Another option:
LM74502, LM74502H Low IQ High Side Switch Controller with Reverse Polarity and Overvoltage Protection
3.2-V to 65-V input range (3.9-V start-up)
• –65-V input reverse voltage rating

Note: This part does not have reverse current protection...

[Smokey]

Here is an app note from TI about ideal diode controllers.  Good comparison between the controllers and the discrete solutions (both P and N): https://www.ti.com/lit/an/slvae57b/slvae57b.pdf

[BillyO]

Sigh!

I do have a 21 page schematic that only uses a mere 321 devices, but I thought the simplicity of the solution would draw ridicule,  So, no.

Ahh, diode?