How to protect power supply against applying reverse voltage?
Reverse voltage
- Thread starter jitenchav234
- Start date
- Status
Couple of diodes would probably be the easiest, cheapest way.
Get yourself a one-way check valve. Voltage goes through but can't come back out.
Diodes
Are you looking for DC reverse polarity protection? Fuse and diode is cheap and effective.
![[Linked Image]](http://www.ozqrp.com/graphics/rev_pol_fuse1.png "[Linked Image]")
OldSparks has the most common method. You can also implement a method wherein the polarity of input doesn't matter, as long as you can tolerate a loss of voltage of two forward biased diodes and the internal ground doesn't need to match the Supply Ground:
Originally Posted by das_peikko
Get yourself a one-way check valve. Voltage goes through but can't come back out.
That's what a diode is.
Regarding to how to protect against it, you need to also have a capacitor or else if the current has nowhere to go, voltage may spike up high enough to fry stuff.
If the other side is just resistance load that's fine, but it wouldn't have generated reverse load. If the other side is inductance load (inductor, motor in regen) then that load can really fry stuff quick, depends on how big it is. I once worked on 2 motor controllers, 1 had 8 giant capacitors inside taking up 1/2 of the space and a cheap power supply, and the other had none what so ever but a nice switching power supply. Guess which one blew up all the time?
If you can't put a lot of capacitors there to buffer the power, do not use switching power supply, use a non-switching one instead (they are more tolerance of reverse and may send it back to the supply side), you will likely drop efficiency from 90% to around 60% easily though.
Get yourself a one-way check valve. Voltage goes through but can't come back out.
That's what a diode is.
Regarding to how to protect against it, you need to also have a capacitor or else if the current has nowhere to go, voltage may spike up high enough to fry stuff.
If the other side is just resistance load that's fine, but it wouldn't have generated reverse load. If the other side is inductance load (inductor, motor in regen) then that load can really fry stuff quick, depends on how big it is. I once worked on 2 motor controllers, 1 had 8 giant capacitors inside taking up 1/2 of the space and a cheap power supply, and the other had none what so ever but a nice switching power supply. Guess which one blew up all the time?
If you can't put a lot of capacitors there to buffer the power, do not use switching power supply, use a non-switching one instead (they are more tolerance of reverse and may send it back to the supply side), you will likely drop efficiency from 90% to around 60% easily though.
Last edited:
- Status
