Negative ground vs Positive ground wiring - here we go again.

[Perrow]

The math works both ways, but we have to keep two pictures in our heads at the same time.

Fortunately, we are capable of doing this if we try.

To paraphrase Friends; "R.G is a girl!" 

[Gurner]

Remember Reinhold Niebuhr's serenity prayer?

God, grant me the serenity
To accept the things I cannot change;
Courage to change the things I can;
And wisdom to know the difference.

Whilst it's always a pleasure & joy to see one of my posts flush out some prose, fear not, this actual issue causes me about three femtoseconds of pondering once every 5 years or so.

[Matt Biermann]

carrejans/govmt lacky:

Allow me to jump in here. Someone please correct me if I am wrong in my explanation. The PNP circuit adds a negative bias to the AC component from your guitar and operates below 0 V ground, hence the "positive-ground" reference point of operation. This signal then has that DC bias removed by the output coupling cap and the AC component is just a "normal" AC waveform cycling around a 0 V center reference point again as it travels to the NPN device via the signal cable. The NPN circuit then adds a positive DC bias to the AC component and operates above 0 V ground. At the NPN device's output, the DC bias is again filtered out, leaving just the AC component again. It is the isolation of the DC operating points of each circuit which makes this work. You aren't really shorting anything out when you use the separate, isolated power supplies, or two different batteries to power each circuit. They just enable you to operate at different points in reference to the 0 V ground reference point. This works by making the 0 V ground reference point either "more positive" or "more negative" than the operating point of your circuit, whichever it may be, but ground is still ground and "0 Volts" for both circuits.

I hope that explains it better. It can be pretty confusing. Especially when you try to envision what the actual electrons are doing during the entire AC cycle at each stage and within each bias situation.

[edvard]

http://www.muzique.com/lab/fuzzface.htm

[Govmnt_Lacky]

Ok. Like Im a 6 year old... 

Lets say I have a PNP Positive ground Fuzz Face and I have a TS-808.

I have NO plugs in the Input or Output jacks.

I plug BOTH pedals into the same DC power supply.

Will I short out my PS even though the pedals are not connected by an audio cable?

[edvard]

Probably not.

If your FuzzFace is TRUE positive ground (the sleeve connection of your guitar cable is connected to +9V) then the guitar cable will be shorting + to - plain and simple.
Without the cable, the two circuits are happily living in their own little worlds.

The solution would be to do like the AMZ article says, so all sleeves are connected to the negative pole.

[Govmnt_Lacky]

Probably not.

If your FuzzFace is TRUE positive ground (the sleeve connection of your guitar cable is connected to +9V) then the guitar cable will be shorting + to - plain and simple.
Without the cable, the two circuits are happily living in their own little worlds.

The solution would be to do like the AMZ article says, so all sleeves are connected to the negative pole.

I proposed that earlier in this post I believe.

On the Fuzz Face, run your +V from the DC jack ONLY (no battery)... to the ground rail of the board and to the 3PDT (for the LED). Then connect your -V to your Input ring, Output ring, LED (via resistor), DC jack -V, and the -V input to the board.

This way, the +V is still getting to the ground rail as it does in the "normal" positive ground circuit BUT..... you now have your input/output jack rings connected to -V.

[edvard]

I proposed that earlier in this post I believe.

On the Fuzz Face, run your +V from the DC jack ONLY (no battery)... to the ground rail of the board and to the 3PDT (for the LED). Then connect your -V to your Input ring, Output ring, LED (via resistor), DC jack -V, and the -V input to the board.

This way, the +V is still getting to the ground rail as it does in the "normal" positive ground circuit BUT..... you now have your input/output jack rings connected to -V.

And you were correct (IMHO), you could even include the battery so long as it was switched out by the DC jack, but that's a given.

I think we've all muddled the whole thought process.
We've all been conditioned to associate "negative" with "ground" from staring at "normal" schematics for days/weeks/years on end.
Then we get this crazy idea to build a "proper" Fuzz Face with it's quirky PNP circuit, and we have to somehow flip our thinking.
So we call things "positive ground" and "negative supply" in an attempt to get our head around it and end up more confused.

I propose we think of two different worlds: Circuit World and Outside World, and propose a few rules for how to interact between them:

The boundary of Circuit world starts at the first semiconductor entry point and ends at the last semiconductor exit point.
Outside world consists of all the various jacks, cables and switches (hardware and semiconductor) that direct Signals into or around Circuit World.
In Circuit World, there is no such thing as "Ground", only Negative and Positive.
When working with Op-amps, there is "Ground", but it is simply a point halfway between Negative and Positive.
In Outside World, "Ground" does exist and by necessity it is always connected to Circuit World's Negative lead.
Signals traveling between Circuit and Outside Worlds must pass through a capacitor upon entering and leaving.

The last problem being Oscillation, which R.G. and others have described, which in most cases can be cured by proper application of a power supply decoupling capacitor.
But barring that, whaddya think?

[Govmnt_Lacky]

But barring that, whaddya think?

I think that this is the exact same thing that I have been getting beaten up about for a few threads now! 

I like your Circuit World/Outside World terminology! 

That was actually the FIRST post that made 100% sense to me regarding the whole Negative ground/Positive ground debate!

According to your terminology... and my idea about wiring... A Fuzz Face would still get the same inputs to "Circuit World" however, the inputs to "Outside World" would change so that a common power supply would NOT short out when this "Positive Ground" Fuzz Face is connected to another "Negative Ground" pedal.

[R.G.]

The last problem being Oscillation, which R.G. and others have described, which in most cases can be cured by proper application of a power supply decoupling capacitor.

The thing that ruined that "most cases... cured... proper application..." is that I found some instances where no amount of proper, well thought out, experienced, skilled power supply decoupling will fix the oscillation.

Just noting that one has never had a problem with elephants from trampling the roses does not mean that there are NO lawn which need regular doses of elephant repellent to keep the roses healthy. Just because you're paranoid doesn't mean that no one, anywhere is not out to get you. 

[Govmnt_Lacky]

Just because you're paranoid doesn't mean that no one, anywhere is not out to get you. 

Thanks for the input RG. I think you may have the wrong idea. Please don't mistake paranoia for the need or desire for an explanation that actually makes sense to those of us that do not do this for a living. I merely wanted a laymans perspective regarding this idea that I had floating in my head. I meant no disrespect... it just so happens that edvard was able to do in one post, what many have not been able to do in multiple posts!

I now have a better understanding of the PNP positive ground vs. negative ground.

[R.G.]

No biggie. I really was chuckling.

The positive/negative ground is infinitely confusing, and I've been typing much the same stuff for well over a decade now. I guess that ought to incent me to go write a "Technology of Positive Ground" . 

The negative side of a power supply is used for ground so often today that it's common to think of negative as ground. That's the first stop on confusing beginners. Second stop is that it's so very inconvenient to use two supplies (or was: AC power adapters are now so cheap that getting a second $20 supply is often the best idea) that people start looking for tech secrets to "fix" this, and they latch on the first layer of EE knowledge: Hey! Either side can be used for ground, just do the caps and pullups right and it all works. And mostly it does, until the times it doesn't. Many of those can be fixed. But some can't.

It's a confusing area, and each layer of new info has its own gotchas attached.

[edvard]

it just so happens that edvard was able to do in one post, what many have not been able to do in multiple posts!

I am humbled that you found my explanation so helpful; glad to be of service 

Though I do agree with R.G. there are some cases where oscillation simply happens and cannot be cured (Hell, that happens with some "normal" circuits I've cooked up...  ), I think it's just something in the nature of how Fuzz Faces work that makes them act so persnickety.

[Perrow]

Hell, that happens with some "normal" circuits I've cooked up... 

Call it a tone generator and stop debugging 

[R.G.]

I think it's just something in the nature of how Fuzz Faces work that makes them act so persnickety.

My particular counterexamples were not fuzz faces, though.

[PRR]

> I found some instances where no amount of proper, well thought out, experienced, skilled power supply decoupling will fix the oscillation.

If you have built a good oscillator, better supply decoupling will improve the oscillation.

Amplifiers oscillate. Oscillators don't. Murphy never sleeps. Corgis keep elephants out of the roses, but won't keep Murphy out of the electronics.

[R.G.]

> I found some instances where no amount of proper, well thought out, experienced, skilled power supply decoupling will fix the oscillation.

If you have built a good oscillator, better supply decoupling will improve the oscillation.

Amplifiers oscillate. Oscillators don't. Murphy never sleeps. Corgis keep elephants out of the roses, but won't keep Murphy out of the electronics.

Yeah. Murphy and I are old friends.

What still bothers me about this issue is that undoing the conversion to negative ground stopped the oscillations. Period. Reconversion started it. No amount of cleaning - or dirtying - the power supply impedance cured it. Makes no theoretical sense. Been through the theoretical and practical issues in depth, including consulting some guys I think are much better at electronics, including RF design, than I am.

I eventually marked that down as "don't do that". It's still an open issue with me.