Author Topic: Reverse polarity protection using mosfets  (Read 1284 times)

0 Members and 1 Guest are viewing this topic.

Offline derGoldstein

  • Regular Contributor
  • *
  • Posts: 146
  • Country: il
    • RapidFlux
Reverse polarity protection using mosfets
« on: March 22, 2017, 07:45:43 am »
I have a circuit with an input voltage of between 5V to 20V, which I want to add reverse-polarity protection to. Attached is the circuit that I'm going to try.
If the power is connected correctly the mosfet allows the current through and the led on the right (which is green) turns on. If it's connected in reverse then the current is blocked and the led on the left (which is red) turns on.

Firstly -- will this work the way I described above?

Second, I want to make sure that I'm using the right mosfet. What I'm looking for is a logic-level mosfet (turns completely on by the time Vgs reaches 5V) which can also tolerate Vgs going up to 20V+. Here are some that I've found:
http://www.mouser.co.il/ProductDetail/Diodes-Incorporated/DMG7430LFG-7/?qs=sGAEpiMZZMshyDBzk1%2fWi4P4wC6csLMYIC%252bvOe09BV0%3d
http://www.mouser.co.il/ProductDetail/Diodes-Incorporated/DMN3018SSS-13/?qs=sGAEpiMZZMshyDBzk1%2fWi4P4wC6csLMY9iJ21xVc378%3d
http://www.mouser.co.il/Search/ProductDetail.aspx?R=BSC100N06LS3_Gvirtualkey63830000virtualkey726-BSC100N06LS3G
http://www.mouser.co.il/ProductDetail/ROHM-Semiconductor/RQ3E100BNTB/?qs=sGAEpiMZZMshyDBzk1%2fWi5brl1xcCtmLSztafmJM%252bzn0STSOkcaEpQ%3d%3d

The maximum current that will flow into the circuit is around 5A, but I want to avoid heat buildup so I looked for mosfets that can handle over 10A, though I understand that some of those are going to be overkill.

Lastly, is there a reason to use a p-channel mosfet instead? Many example circuits opt for a p-channel mosfet connected to the positive input rather than an n-channel mosfet connected to the negative input. But n-channel mosfets are a lot cheaper and there's a much wider range of them. Is using a p-channel mosfet more efficient in some way?

Thanks in advance.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: Reverse polarity protection using mosfets
« Reply #1 on: March 22, 2017, 08:05:16 am »
(a) Many LEDs have an absolute maximum reverse voltage rating of only 5V or so.  Applying more than that will possibly damage the LED, so consider not using the reverse polarity indication LED like that.

(b) MOSFETs rated at 10A or any number continuous drain current are usually rated that way with AT LEAST the assumption of a particular heatsink which must be present or a better one must exist.  Even more usually that is specified at a particular abmient temperature, so derate accordingly.  Even more usually the continuous drain current limit is specified at IMPOSSIBLE or UNREALISTIC levels of heatsinking, e.g. the case temperature being 25C no matter how much heat the MOSFET is dissipating.  So even a 10A rated part MIGHT be not conservative enough for your circuit if 5A is a commonly applicable use case and maybe you will operate at high temperatures or with poor heat dissipation.  But probably it is OK with a 4:1 power difference between 10A and 5A by I^2 * R.

(c) Usually PMOS FETs are used because usually the (-) is the supply return but ALSO is GROUND and the (+) is a supply line.  So if the high side is not blocked in the case of desiring to switch off the power (e.g. due to a supply reversal) the whole circuit can "float" at the (+) supply line potential.  Now consider what happens if any other part of the circuit is connected to GROUND like perhaps some video or audio or other connection other than the PSU (-) lead.  Then the circuit will have PSU (+) as well as (some other GROUND) connected.  And it will be subject to the power supply with correct or incorrect polarity despite the broken PSU (-) connection.  If your circuit is always floating from the PSU / earth maybe you can block the low side.

(d) You show two GND connections that if literally made will short out the FET.

Otherwise... yeah I guess, I didn't look at the FET data sheets, just the circuit diagram.

PCH usually has higher RDSON for a given internal silicon size of fet than NCH so actually NCH is more "efficient" it is just harder to use in a high side switch and as I said not all applications toerate low side switching due to wanting GROUND to be GROUND and not something maybe somewhat different than ground.




I have a circuit with an input voltage of between 5V to 20V, which I want to add reverse-polarity protection to. Attached is the circuit that I'm going to try.
If the power is connected correctly the mosfet allows the current through and the led on the right (which is green) turns on. If it's connected in reverse then the current is blocked and the led on the left (which is red) turns on.

Firstly -- will this work the way I described above?

Second, I want to make sure that I'm using the right mosfet. What I'm looking for is a logic-level mosfet (turns completely on by the time Vgs reaches 5V) which can also tolerate Vgs going up to 20V+. Here are some that I've found:
http://www.mouser.co.il/ProductDetail/Diodes-Incorporated/DMG7430LFG-7/?qs=sGAEpiMZZMshyDBzk1%2fWi4P4wC6csLMYIC%252bvOe09BV0%3d
http://www.mouser.co.il/ProductDetail/Diodes-Incorporated/DMN3018SSS-13/?qs=sGAEpiMZZMshyDBzk1%2fWi4P4wC6csLMY9iJ21xVc378%3d
http://www.mouser.co.il/Search/ProductDetail.aspx?R=BSC100N06LS3_Gvirtualkey63830000virtualkey726-BSC100N06LS3G
http://www.mouser.co.il/ProductDetail/ROHM-Semiconductor/RQ3E100BNTB/?qs=sGAEpiMZZMshyDBzk1%2fWi5brl1xcCtmLSztafmJM%252bzn0STSOkcaEpQ%3d%3d

The maximum current that will flow into the circuit is around 5A, but I want to avoid heat buildup so I looked for mosfets that can handle over 10A, though I understand that some of those are going to be overkill.

Lastly, is there a reason to use a p-channel mosfet instead? Many example circuits opt for a p-channel mosfet connected to the positive input rather than an n-channel mosfet connected to the negative input. But n-channel mosfets are a lot cheaper and there's a much wider range of them. Is using a p-channel mosfet more efficient in some way?

Thanks in advance.
 
The following users thanked this post: derGoldstein

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: Reverse polarity protection using mosfets
« Reply #2 on: March 22, 2017, 08:07:47 am »
BTW LTSPICE is usually a good way to check / prove circuit function assuming that you actually measure / apply realistic values at every point in the circuit since it will happily let you have near infinite voltages or currents or whatever and see no problem, and smoke doesn't come out when you apply reverse polarity.  So you have to check with a "meter" etc.  And it has no real concept of temperature or breakdown voltage.
But it'll tell you when you have ample current flowing vs. insignificant.


 
The following users thanked this post: derGoldstein


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf