Question about positive grounding

[the_creature]

Hey!
I have a question about positive grounding.
I've just built myself a PNP Fuzzface which obviously requires positive grounding, and everything works and sounds awesome!
but i've only used it as a lone pedal. so my question basically is, how does this other pedals in the chain if one has positive grounded input and output? so i am not talking about the powersupply, just the Mono and the stereo jack. because i mean the ground is connected from one pedal to another through the patch cable.

[bluebunny]

They will all play nicely together.  Ground is ground.

i am not talking about the powersupply

It's good that you recognise this.  Just take care when you do start to think about power supplies.  Your Fuzz Face will need an isolated supply (or a battery).

...and welcome!

[induction]

how does this other pedals in the chain if one has positive grounded input and output? so i am not talking about the powersupply, just the Mono and the stereo jack. because i mean the ground is connected from one pedal to another through the patch cable.

The difference between positive and negative ground is in the voltage of the supply rail, not the ground voltage, which is zero by definition. "Positive ground circuit" means that ground is more positive than the supply rail. It could easily be called a "negative supply rail circuit". Likewise a negative ground circuit has a ground point that is more negative than the supply rail. In either case, ground is 0V because both adapters and batteries don't supply absolute voltages. Instead, they supply differences in voltage between the two leads. Those voltages are isolated from the mains voltage, so they don't take on specific values until one of them is tied to a known reference voltage. In this case, the known voltage is provided by the ground connection at the input and output jacks. There is no problem connecting positive and negative ground circuits together as long as you don't daisy chain them. Without a daisy chain, each pedal is only connected at the signal voltage and ground voltage nodes. Your delay pedal has no way of knowing what supply voltage you used in your fuzz face, nor any of the other DC voltages inside the circuit.

You may be interested to read another recent thread about this issue.

[antonis]

There is no problem connecting positive and negative ground circuits together as long as you don't daisy chain them.

Although it should be an overkill for a FuzzFace, it should be an anxiolytic idea to incorporate a small voltage inverter (e.g. 555 based) and not worry anymore about daisy chain's compatible polarity..

[GGBB]

Although it should be an overkill for a FuzzFace, it should be an anxiolytic idea to incorporate a small voltage inverter (e.g. 555 based) and not worry anymore about daisy chain's compatible polarity..

Even easier - just attach the power supply differently:

[R.G.]

Even easier - just attach the power supply differently:

That is easier, OK.

But it does not always work, for reasons that are difficult for beginners to deal with. In many cases, it will give you noise and/or oscillation that cannot be cured except by putting the power supply back the way it was originally. Worse yet, the technical reasons why this is so are not clear.

It is not a good thing for beginners to deal with, and I do not recommend it. Sure, sometimes it works. Until it doesn't. 

[GGBB]

Jack addresses that issue in the accompanying article and concludes "If for some reason oscillation does occur, a quality low esr capacitor from the positive supply to ground will solve the problem everytime since it effectively places the power rails at the same ac potential." But he also closes with "The circuit can easily be converted back to positive ground if you get oscillations that are not cured by the power supply filter."

[R.G.]

And that is indeed one of the possible fixes. The problem is that it works sometimes, but not every time, just like all the other fixes I've tried or had suggested over the years.

I ran into this issue back in the late 1990s. I had a pedal to fix that was positive ground and happily reversed the power supply, just like the baby-steps theory of power supplies suggests would work. It oscillated.

No problem, I thought. I'll just use some power supply decoupling caps. It oscillated differently.

Well, I thought, it changed, so power supply impedance is at least involved. I'll just work at it harder. To make a long, sad story shorter, nothing I did fixed the issue. Re-reversing the power supply to the normal positive ground fixed it first time, every time.

In short, no, a quality low-ESR capacitor will not solve the problem every time. It will help in many case, and may fix some of them, but there are cases where the solution is simple: put it back to positive ground. Some circuits are simply fussy about their grounding. I suspect that there is not one cause for this, but many. And while analog circuit theory does say that simply reversing the power supply grounding works, in practice, Mother Nature is more devious than that. As Yogi Berra said "In theory, theory and practice are the same. In practice they aren't."

On the other hand, this is a slippery issue. Simply reversing power and ground works often enough that people keep trying it and it works mostly, no muss, no fuss - until it doesn't. Then using power filtering, rearranging wires, reworking the ground paths, any of a number of mystical, esoteric circuit tricks get applied, and some of them work on some cases, so the simplistic response is to declare victory and post that recipe for victory to the internet.

Then the next beginner tries it with a hard case and none of the recipes work. The usual process here is to blame the beginner for doing the fix wrong. But reversing the power supplies back to positive ground does, in fact, always work, even for the hard cases.

I personally hate it that I have not been able to find out the underlying cause of the problem and publish a simple fix, after running into the issue over a decade ago, and seeing a series of people posting about reversing positive ground and having oscillation. I'd like to think I can go dig out the technical reasons for why something happens in a circuit, but I've been disabused of that conceit for a long time. So for this one, the advice to beginners is "don't do that". More advanced folks that can deal with the consequences can mess with the plethora of band-aids and some of them will work, in some cases.

But there isn't a single slam dunk that I know of, only partial fixes. And I have looked.

[jaybrau]

There are only two differences in the switched circuit:

• The signal now oscillates relative to ground instead of +9V.
• The more negative rail is now connected to the ground shared by the rest of the pedals, amp and guitar, where the more positive rail had been before.

It doesn't seem that the relative swing would matter as long as the input has a coupling cap, so I would guess the oscillation would then be caused by the introduction of ground loops.  Do you have an example of a circuit where reversal causes oscillation?  Reversal does work on a stock PNP Rangemaster circuit, and doesn't cause oscillation.

[R.G.]

It doesn't seem that the relative swing would matter as long as the input has a coupling cap, so I would guess the oscillation would then be caused by the introduction of ground loops. 

My best theoretical explanation is the introduction of shared ground impedance, which is of course the origin of ground loops. And tinkering with ground and power supply impedance does affect the situation. And low impedance filter caps and so on do help many cases, but not all.

Do you have an example of a circuit where reversal causes oscillation?

Unfortunately, the things that cause problems with ground/power reversal don't show up on schematics. They are layout and wiring dependent. In general, the higher the gain and the higher the currents involved, the more likely this gets. But that's true of all grounding and parasitic oscillation situations, too.

There are multiple posts in this forum over the years of someone setting up a Fuzz Face or one of the buffered-input fuzz face types and getting oscillation with reversed ground. Reverting to positive ground fixed them all.

Reversal does work on a stock PNP Rangemaster circuit, and doesn't cause oscillation.

the Rangemaster circuit is low gain, low current and relatively low impedance, so it's at the low end of susceptibility. That does not mean that there isn't some Rangemaster wiring and setup situation that would oscillate with reversed ground.

As I noted, there are are two kinds of people who tinker with reversed ground circuits; those that have had it oscillate, and those that have not *yet* had one oscillate. If someone wants to play with reversed ground, and is knowledgeable enough to deal with it when and if it does, fine. Otherwise, they're likely to turn up here saying that their new reversed-ground [whatever] oscillates. Getting away with it is fine - until you don't.

[Tony Forestiere]

If someone wants to play with reversed ground, and is knowledgeable enough to deal with it when and if it does, fine...Getting away with it is fine - until you don't.

I hear "Adrenaline Junkies only need apply". 

[johngreene]

This explains the reason for the only oscillations in a fuzz face circuit that I have seen.

FF oscillation