another polarity protection question

[flintstoned]

I've read R.G.s article and searched lots of posts, but now I've  caved and posted because I didn't see anything come right out and answer my assumption.  Just wondering if the series polarity protection diode makes the shunt diode obsolete?  Since I use a one spot exclusively and don't bother adding the battery wiring, I don't need to be concerned about the voltage drop that comes with the series diode. But I wanna start making my pedals idiot proof (aka drummer proof!) I'd like to protect against reverse polarity and AC power supplies of death. Just wanna be positive (ugh, sorry, didnt mean that) about this.

[R.G.]

A series diode is always certain polarity protection. As long as you can afford the loss of voltage that the diode drop gives you, you're fine. The diode voltage loss is the only reason not to use the series diode setup for polarity protection in the limited milieu of effects pedals at least.

[Fender3D]

An idiot proof solution:
you may use a diodes bridge instead of a single diode
connect the 2 ~ tips to the pwr jack,
the - tip to gnd
and the + tip to +V PCB connection
(assuming you have a negative ground circuit, reverse the connections in case of positive ground)
this way you'll never concern about reverse polarity or AC supply anymore.

[earthtonesaudio]

The diode voltage loss is the only reason not to use the series diode setup for polarity protection in the limited milieu of effects pedals at least.

In a few specific effects pedals you might also wish to consider the power dissipation in the diode (for example, tube-based pedals which draw high current), and choose a diode that can take the heat.

but like RG said, most times you don't need to worry about it because the power draw is so low.

[caress]

i always use a 1N5818 in series with the supply, as it has a less of a drop than a 1N4001.

[flintstoned]

Thanks y'all!

[R.G.]

An idiot proof solution:
you may use a diodes bridge instead of a single diode
connect the 2 ~ tips to the pwr jack,
the - tip to gnd
and the + tip to +V PCB connection
(assuming you have a negative ground circuit, reverse the connections in case of positive ground)
this way you'll never concern about reverse polarity or AC supply anymore.

...until you hook it up to play in a pedalboard. The diodes on the bottom side move the signal ground one diode (however small that is) off the minus side of the power supply running it. If that gets hooked to another pedal you now have a ground offset of one diode drop, and the voltage varies with how much current runs through it. It works OK if all your pedals have separate supplies, or they all have the same supply. In the middle of that will potentially cause problems.

I worked on making a diode bridge power input for a long time, never reached a solution I liked with it. The best was with active devices instead of diodes making up a bridge. An active bridge can get much lower voltage drop than a simple diode. Both bipolars and MOSFETs work. If properly driven, you can get quite low drops. But it's hard to get things quiet when there's a voltage offset in the ground side.

A simple MOSFET polarity protector or the "cheap - and good - polarity protection" at GEO give you the advantages of series protection without losing a diode drop in the + line.

[Fender3D]

... In the middle of that will potentially cause problems...

What if you connect the battery after the bridge?
Won't the 2 diodes protect from reversing it being connected in series among them and in parallel with battery, and no diode drop on battery?

[R.G.]

What if you connect the battery after the bridge?
Won't the 2 diodes protect from reversing it being connected in series among them and in parallel with battery, and no diode drop on battery?

Sure. As long as all you're considering is the one pedal. But players haven't used only one pedal for decades now. And more people don't use batteries at all.

The issue is how are all the other pedals connected to ground and each other too. There are many ways to introduce ground offsets and noise, and deliberately introducing a diode drop offset in the minus side of the power supply, which is what mostly passes for ground, is a good way to get things wrong. I'm not saying it can't be made to work, just that there are pitfalls there, and with more pedals, more pitfalls are available.

The diode bridge in front of the pedal is a neat idea. I discussed this with aron years ago. I could never find a simple, cheap, and foolproof way of guaranteeing and foolproofing the interconnection problem. An active MOSFET bridge to minimize the "diode drops" down to millivolts was as close as I got. That is in general over the heads of most DIYers.

[Ice-9]

And interesting little article on MOSFET rectification and how it works can be read here,

http://www.thetaeng.com/FETBridge.htm