Cheap Polarity Protection

[FrankB]

Of course, none of us would ever accidently hook up a DC powered device and reverse the polarity feeding it. Nope, never.

So here is the solution.
Add in a bridge rectifier to the input. Just use a bridge with the correct amperage and voltage or higher; or 4 diodes to make the bridge.

That way, if you reverse polarity feed the DC going in, it won't fry the device.

Yes, there is a small voltage drop in the bridge, but most of the time, it won't make much difference in the operation.

[Peter.N.]

Good idea but the voltage drop will be in the order of 1.2 volts (two diodes in series), OK if the voltage is not critical but not good if it is.

Peter

[camtechman]

I do electronic work for members of a local model railway club and, now that many of them want mods done to trains & equipment, such as fitting LEDS in place of the painted running lights (front & rear), also bright white LED headlights (front & back),and that the trains are run in either direction (reversing the polarity of the power supply), fitting a bridge rec is essential to prevent damage to the electronic components.

[kalee20]

I was tidying up a PA amplifier recently, mains / 12V operation. There was no reverse polarity protection so I used an inverse-parallel single diode (MR751) across the supply input, and an upstream fuse (with a couple of spare fuses in a bag taped to the supply leads).

I didn't fancy the voltage drop of a series diode, much less a bridge!

[GMB]

Yes, the voltage drop can be a big issue especially in anything powerful (and hence more likely to die in the event of a problem).

All the military receivers that have such protection that I have seen do it using a fuse and a high-power zener diode. This gives over-voltage protection too!

[Radio Wrangler]

Another technique is having a relay to connect the power, and having the coil powered via a small diode. If the polarity is wrong, the relay won't pull in. This can be good for very low voltage drop and very high currents where the coil power isn't significant.

David

[Nickthedentist]

I like that, David.

[TrevorG3VLF]

A Schottky diode will drop less voltage than an ordinary silicon diode and a power FET can be used to drop a negligible voltage.

The FET is used the wrong way up so the internal diode conducts and then the gate will turn the FET on giving a resistance of less than 0.1 ohm. You will need a supply of 10V or so.

[cmjones01]

The other issue with a bridge is that it leaves a diode voltage drop between the internal negative rail of the equipment and the battery negative terminal. This can be a serious problem if battery negative is assumed to be 'earth' and the equipment (such as a PA amplifier) also uses its negative power supply terminal as 'earth'. You end up with a sort of earth loop on steroids, with a diode in it, which can have nasty effects.

Chris

['LIVEWIRE?']

Some Car radios in the 1970s were produced with a an 'automatic polarity selector', which was a bridge recitfier connected as Frank B. suggests. These were all, AFAIR, mono radios with maybe 5watts audio o/p at most, and used 4 x 1N400x series diodes. Obviously a drop of ca. 1.2v between the supply and the radio circuittry exsisted because of the diodes, which meant that it was being run from about 10 to 13v, assuming a battery voltage of 11-14.4v, but no problems arose as a result, other than slightly lower o/p power.

[Peter.N.]

The 'crowbar' circuit (zener diode across the supply) was used in most of the later CRT TVs as a protection against excessive HT but will work equally well with an ordinary fairly heavy diode and a fuse in the circuit, although the diode often goes s/c it still acts as a protection. I have also seen the same circuit operating a relay, not quite so brutal but not quite as fast either.

Peter

[Skywave]

Bridge: 1.2 v. drop; single diode: 0.6 v. Even less for a Schottky diode.

Al.

[G6Tanuki]

The relay-with-diode-in-series-with-the-coil is my preferred approach; it allows optimally-short power-lead wiring to minimise voltage-losses, and also removes the need for the on/off switch to have a significant current-rating.

It was the standard approach back in the days of valve/transistor 'hybrid' mobile-radio gear [Pye Cambridges/Vanguards etc] and is still used on a lot of higher-power commercial two-way gear.

[Peter.N.]

I used to have a PMR system with Cambridge and Vanguard sets. I never connected one up backwards so can't give any test results.

Peter

There is also the problem of getting it wrong in the first place, guaranteed blow up every time!

[TrevorG3VLF]

This is how a fet can be used for polarity protection with negligible voltage drop.
You will need to chose a fet which can be turned on with the voltage you have.
Some gate protection may be desirable

[FrankB]

I installed the bridges in many car radios back in the 70's, as the teens used to hook up the repaired radios backwards right after I fixed them, blowing the AN214Q's in the KP 500 Pioneers and the DS 501's & DS 503's in the Delcos' . We got tired of the expense and time to re-repair the radios. (I worked at a Delco and Pioneer ASC repair shop).
It cut call backs to > .01%! And never a complaint in over 300 radios I did this to.
(A simple adj. of the tape motor speed control pot and it was running the correct speed. It took under 10 minutes to do the modification.)

I realize some devices balked at having less than full voltage on them, however the bridges cost like 5 cents each back then in lots of 1K from the local surplus store, and Zener's were way more expensive. No room in the radios for a relay either. We tried many things and this was the cheapest and easiest at the time.