Reverse polarity protection options.

Hi, Brad,

What about a really simple solution like a diode and a fuse? Put the diode across the supply,an put the fuse in series. If the leads get reversed, the diode will conduct, blowing the fuse.

-d2

Actually, this is what I came up with on the fly and it works perfectly.  It's a discrete module, not to be incorporated onto the final board, used only for protection against bench-testing errors.

It's simple enough and it works quite well.   Each approach is a comprimise.   A diode without fuse inserts voltage loss.  A reversed and paralleled diode (if that is what you are suggesting) with fuse or fusible link, while not inserting loss, requires replacement with every "error".   I also (i am not an EE so I hope someone will weigh in here) am concerned with the implementation of any fuse so simply set up.  In the milliseconds required to burn the fuse the opamp and amplifier chips are exposed to reversed polarity.  According to all I've read, the dedicated reverse polarity ICs from Fairchild or properly implemented Pmos fets from TI are orders of magnitude faster at gating against reversed voltage.

It's sometimes difficult to relate the application in a question so that the answers are most germain to the problem.  I don't need polarized or color coded plugs.  That is just invitation, on the bench, for further error.  A much better solution for simply bench testing various power supplies and boards is the discrete module I imaged above.  Correct polarity and all's fine.   Reverse polarity and the led warns and nothing touches the attached board.  I can use it on the bench as tool for any future similar reverse polarity protection bench testing needs.   But..........

But it's frustrating that I keep finding suggestions for simply/elegantly implemented pmos fets or that IC from fairchild which, for less than two dollars per application, might simply be incorporated right onto the board and left there as permanent reverse polarity protection. No blown fuses.  No servicing. Just a coldly logical gate that says YES or NO to correct polarity without blowing any fuses or inserting any current/voltage loss. As a matter of fact, it appears that the fairchild IC had a pin to power an led warning upon reversed polarity.  

Surely this is the best option. I could simply order half a dozen of those Fairchild chips to experiment with.  Or someone at e2e.ti.com can identify a TI equivalent.   But I've posted in at least a dozen boards so far and have mostly been directed toward the most rudimentary protection involving polarized plugs and color coded plugs despite providing schematics and IC suggestions for better solutions. 

Why is this such a difficult prospect to make recommendations upon?

Well, in my opinion a MOS fet is such a simple solution that I dont really see the need to develop an IC for a function that a single component does very well.

Aren't you reading my posts?

."I keep finding suggestions for simply/elegantly implemented pmos fets or that IC from fairchild which, for less than two dollars per application, might simply be incorporated right onto the board and left there as permanent reverse polarity protection. No blown fuses.  No servicing. Just a coldly logical gate that says YES or NO to the polarity without blowing any fuses or inserting any current/voltage loss."

Your response....

Well, in my opinion a MOS fet is such a simple solution that I dont really see the need to develop an IC for a function that a single component does very well.

Yes, yes... double and triple that!!!!!  Aren't you reading the full thread?  Please, on my knees.. tell me which Mos fet to use in this instance, with regard to the values I've proposed.   That's all I'm really asking here.   It is my  understanding from reading fairly deeply into this yesterday and today that it's an uncommon mosfet that will handle >12vdc on its own.  It needs to be in circuit with another diode and resistor, whose values must be carefully determined according to voltage and amperage in the circuit.  

I swear I will not darken your door with such maniacal  pursuits until mid summer at least if someone will step up and offer a "simple MOS fet" three component circuit to fit my 15 to 29vdc at up to 5 amp present projects.

Sincerly

Brad, 

Most power Mosfets I know of have a max. Vgs of 20V, so if your range is up to 20V you should be good. If the range is up to 29V, or if you are like me and dont want to be too pressed for 20V input being close to the max., you can use a Zener in parallel to Vgs and a 10K resistor to ground. There are many, many devices you can use for your application, for example IRF9317 for Vin max of 20V or IRF7240 for Vin max of 29V. I am sure that if you check other manufacturers, like Vishay, you will find more options. I may be mistaken, but I think that TI Mosfet range of solutions falls a littler short of your specs.

If you have any more questions please dont hesitate to ask.

Best regards

Avi

Hi, Brad,

You might try posting in the forum that I think has apps engineers working on the LM5060 which I believe is here.

-d2

Brad,

Or, better yet, our Automotive forum!!! You can find it here.

-d2

Thanks you guys.  And most sorry for the histrionics.  I was just a little frustrated with what I thought would be the simplest part of this project.  It's ended up occupying more time than I'd bargained for. 

Brad,

No worries. Ye olde 80/20 rule strikes again, eh?

-d2