Polarity Protection? Which pedals need it?

[Canucker]

Which pedals have you found require polarity protection? and which ones don't???  is it as simple as adding a diode between the + lead of your power supply and the circuit board?

[R.G.]

Which pedals have you found require polarity protection? and which ones don't??? 

All pedals need polarity protection as a bit of insurance against human mistakes in connecting power.

Well, OK, all pedals you want to survive accidental reversed leads. If you don't care if it dies when you (or your bass player, or...) connect a backwards power supply, it doesn't need polarity protection.

is it as simple as adding a diode between the + lead of your power supply and the circuit board?

It can be. There are many possible ways.

[R O Tiree]

Have a look at R.G.'s design for cheap - and good - polarity protection

[Canucker]

Have a look at R.G.'s design for cheap - and good - polarity protection

So you need one NPN and one PNP transistor? or is it meant to be an "either/or" situation? Also I'm confused by the arrows on the right of the layout....am I out to lunch or is it drawn out slightly confusing?

[DiscoVlad]

I usually do something like this after having seen it in Electronics Australia or the DSE catalogue back in the '90s... Then being reminded of it by RG:
It uses the (parasitic) body diode of the MOSFET as series protection, but because the transistor is switched on, the voltage drop is limited by the mosfet's Rds(on) rather than being a fixed 0.6V of a silicon diode.
I'm not sure how well it handles an AC supply being connected, a spice sim shows it shouldn't be a problem, but I'm not game to try it on an actual pedal...



Q7, D3, and R31 are the actual polarity protection, the other parts are for the indicator LED, and power supply filtering.

This has got me thinking, with basic shunt diode protection would adding a polyswitch or similar device (like below) solve the Power supply vs. Diode fight to the death?

[Canucker]

bump!

[Gurner]

This has got me thinking, with basic shunt diode protection would adding a polyswitch or similar device (like below) solve the Power supply vs. Diode fight to the death?

I guess it then boils down to the inherent resistance of the polyswitch (which I have no idea what that typically is), because clearly there'll then be a voltage loss attributed to it.

Also how quickly will it react - resistance wise - to that diode being reversed biased? Fastenough to stop the diode frying?

Iersonally I like the simplicity & low compoinents count of a schottky diode in series with the battery inpunt rail & just suffer/tolerate the small voltage drop.

[DiscoVlad]

I guess it then boils down to the inherent resistance of the polyswitch (which I have no idea what that typically is), because clearly there'll then be a voltage loss attributed to it.

Yes there's a voltage drop (well, current limiting), that's kind of the point.
Under normal operating conditions, their resistance is low (tens of ohms at most) and the current through the device is not enough to trip it.
Under fault conditions (backwards battery) the diode conducts, and lets a large current through... This causes the PPTC to quickly heat up which makes its resistance become very high (Megaohms) - substantially reducing the current through the diode, and preventing it from burning out.

Also how quickly will it react - resistance wise - to that diode being reversed biased? Fastenough to stop the diode frying?

With the right device, certainly.

A quick and dirty search finds: http://www.mouser.com/ProductDetail/TE-Connectivity-Raychem/RXEF005/?qs=hv6pn79dJPQLqv1W3AovI2PV%2f%252b6vvV%252bjKMT%2fsE%2fu4l4%3d
This trips in 10 seconds at its nominal trip current of 100mA, but in less than 0.1s above 500mA. Obviously you'd need to choose an appropriate device for however much current your  pedal circuit draws (In my experience, the indicator LED is the single greatest current draw unless you've using a delay chip or something similarly hungry).

I personally I like the simplicity & low components count of a schottky diode in series with the battery inpunt rail & just suffer/tolerate the small voltage drop.

A series mosfet is 2 or 3 components taking up about 0.2" x 0.3" of space (the diode isn't really necessary, and you can probably also get away with not using the gate resistor) and has <10mV forward voltage drop and unmeasurable reverse leakage current for every circuit I've used it on.

[deadastronaut]

isn't just a 1n4004 from 9v to ground in this case polarity protection?.. .

http://gaussmarkov.net/layouts/drboo/drboo-project.pdf

[Gurner]

I guess it then boils down to the inherent resistance of the polyswitch (which I have no idea what that typically is), because clearly there'll then be a voltage loss attributed to it.

Yes there's a voltage drop (well, current limiting), that's kind of the point.

I understand what it's doing.... what I meant was to me, the main important point is under non-reversed polarity circumstance, how much resistance it has & therefore what it wastes in supply voltage (vs. using schottky instead)

isn't just a 1n4004 from 9v to ground in this case polarity protection?.. .

http://gaussmarkov.net/layouts/drboo/drboo-project.pdf

under reversed supply voltage conditions, that diode is essentially shorting the power +ve to the Power-ve.....at 9V, that's ugly at best, but likely to make the diode turn into powder! (esp if it's not a battery supply, but a wall wart)

[DiscoVlad]

I understand what it's doing.... what I meant was to me, the main important point is under non-reversed polarity circumstance, how much resistance it has & therefore what it wastes in supply voltage (vs. using schottky instead)

The one previously linked has 20 ohms resistance... It'll drop 1V for a pedal drawing 50mA, so probably isn't suitable. However there seem to be plenty of other models with resistance well below 1 Ohm.
Any device with resistance around that, and with trip current less than whatever I(max) for the shunt diode is, would work.

[Canucker]

http://farm5.static.flickr.com/4034/4635974624_e8f3894336_o.jpg are we liking this?

[Skruffyhound]

According to my datasheets although the pinout is good the bodies of the transistors are  faceing the wrong way.
There are often, confusingly, different pinouts for the same component from different manufacturers, just to keep us on our toes.
I haven't seen alternatives for these two though.

[Skruffyhound]

I've just noticed that the PNP transistor is hooked up backwards on this layout ^^^
Emitter should be on Vin and collector Vout

Sorry I didn't pick that up before but I was just cross-referencing a few of these layouts and it jumped out at me.
The NPN is ok, just body facing the wrong way round.