Circuit Theory: NPN/PNP Flipping with Negative Ground

[trad3mark]

Ok, so I'm not great on circuit theory, and the interweb seems to overcomplicate things, so here goes. I've a question that's confusing the hell out of me, but i've diagrams etc to explain my blight. I'm sure it's a very simple yes or no question, but here goes. In advance of this, my scanner is broken, and my diagrams aren't great, but they're quicker than if i was to draw it in a cad programme, so bare with me.

I've noticed in some schematics, they seem to take PNP applications with positive ground and sortof "vertiacally flip" the schematic, still using PNP's, but change to negative ground. A good example of this is the Fuzz Face, and second section of a Fuzz Factory. on a related note, there's the way the NPN version of the Fuzz Face goes from positive ground to negative ground by changing NPN's to PNP's and swapping around the polarity of the electrolytes.

So this got me curious as to how I could do this with an old favourite of mine, the Big Muff. Now i know that I could swap in PNP's and change to positive ground, but i want to IMPERATIVELY keep the negative ground. So how do I go about changing over to use PNP's? For your benefit, I've a diagram! A whole one!!!



So this diagram shows two almost identical clipping stages from a Muff. The difference? well the second one uses a PNP. Everything else is the same. The question is, amidoinitrite? If i want to keep the negative ground, is it as easy as that, or, do i have to go all fuzz face on this and flip it so that the 120R and 100k of the PNP go to +ve and the 10k resistor goes to ground? I was going to do that, but then I realsied that in the case of the FF/FF, just about everything flips, and they all seem to cancel eachother out or something.

Help?

[R.G.]

I've noticed in some schematics, they seem to take PNP applications with positive ground and sortof "vertiacally flip" the schematic, still using PNP's, but change to negative ground. A good example of this is the Fuzz Face, and second section of a Fuzz Factory. on a related note, there's the way the NPN version of the Fuzz Face goes from positive ground to negative ground by changing NPN's to PNP's and swapping around the polarity of the electrolytes.

So this got me curious as to how I could do this with an old favourite of mine, the Big Muff. Now i know that I could swap in PNP's and change to positive ground, but i want to IMPERATIVELY keep the negative ground. So how do I go about changing over to use PNP's?

You're setting yourself up for problems.

The theory says that an ideal power supply is a source of constant DC voltage and is a true short circuit for all AC signals. That is, the circuit does not care whether you ground the + side or the - side of the power supply, because both + and - look like a short circuit for AC. In the real world, there are no ideal power supplies, only ones which are closer or farther away from the ideal. This matters because the inside of the circuit depends on one or the other of + or - to be rock solid and zero impedance as a reference for signals. Using the opposite side introduces the power supply imperfections between the reference for signals and the power drains of the circuit. That is, you're confusing the circuit by telling it that "ground" has voltage wobbles on it that are a reflection of its own action.

This sets you up for various circuit malfunctions, including noise, intermittent noise, and full-bore unstoppable oscillation. That's the bad side. It is also possible that you'll get away with it. But it is getting away with it.

For your benefit, I've a diagram! A whole one!!!

Just a bit of nit picking - that's not the correct schemo for the BMP clipping stage. But I know what you mean.

So this diagram shows two almost identical clipping stages from a Muff. The difference? well the second one uses a PNP. Everything else is the same. The question is, amidoinitrite? If i want to keep the negative ground, is it as easy as that, or, do i have to go all fuzz face on this and flip it so that the 120R and 100k of the PNP go to +ve and the 10k resistor goes to ground? I was going to do that, but then I realsied that in the case of the FF/FF, just about everything flips, and they all seem to cancel eachother out or something.

You need to flip the entire circuit upside down.

To make this illustration work and bias properly, you need to
(1) flip the PNP vertically so its collector goes to the 10K/470K and its emitter to the 120 ohm.
(2) cut out the whole PNP circuit from 10K resistor load all the way down to the ground symbols, and flip THAT whole thing vertically.

[trad3mark]

Cheers R.G.! Massive help with the theory side. And I know what you mean about it not the correct schemo, but yeah, close enough for the moment.

You need to flip the entire circuit upside down.

To make this illustration work and bias properly, you need to
(1) flip the PNP vertically so its collector goes to the 10K/470K and its emitter to the 120 ohm.
(2) cut out the whole PNP circuit from 10K resistor load all the way down to the ground symbols, and flip THAT whole thing vertically.

ok, so i'm a little confused. I get the bit about flipping the PNP so the collector goes to the 10k/470k and the emitter to 120R. But then part 2, does that mean that it's like this:
(The one on the left is the original/semi-original/NPN version. The one on the right is the "flippedPNP version", with the top part, being where +9V would be. It got cut off...)

[R.G.]

ok, so i'm a little confused. I get the bit about flipping the PNP so the collector goes to the 10k/470k and the emitter to 120R. But then part 2, does that mean that it's like this:
(The one on the left is the original/semi-original/NPN version. The one on the right is the "flippedPNP version", with the top part, being where +9V would be. It got cut off...)

Yes.

The problem with this is as follows: everywhere you have a ground/earth symbol, insert a resistor and an inductor. All of those meet at one place that's the real, true, no-fooling ground that you can never get to. There's another resistor and inductor in series between the + power supply and every place that supposedly connects to it. These resistances and inductances are made up of the real copper and other metals the circuit traces and wires are made from. Unfortunately, you have to connect your incoming signal ground lead to somewhere. If it happens to be at a "ground" that has currents through it that make it different from "real" ground, and you circuit has enough gain, it takes the voltages made by the current flows as input too.

It's a bad enough problem with only copper wires between places that are supposed to be "ground"; when you put "ground" as the circuit sees it on the other side of the battery or power supply, it gets worse. How much worse depends on everything in the circuit.

Many times you get away with it. Sometimes you don't.

[trad3mark]

this is sounding way more complicated than i had initially hoped. I thought it was a matter of doing the above, and being able to make a PNP based muff, without having to change the polarity of everything. Doesn't look like i'd get away with it.

Might be worth my while to breadboard part of it, even just the opening gain stage to see what it's like though. If even just out of curiosity.

[R.G.]

this is sounding way more complicated than i had initially hoped. I thought it was a matter of doing the above, and being able to make a PNP based muff, without having to change the polarity of everything. Doesn't look like i'd get away with it.

You might. It actually happens that you get away with it more often than not. But knowing the problems lets you be aware of what you're getting into.

Might be worth my while to breadboard part of it, even just the opening gain stage to see what it's like though. If even just out of curiosity.

If you have the time and motivation, go for it. Be aware that since the parasitic stuff, the resistances of the contacts and wires are what make it not exactly like the theory, that a breadboard may be better or worse than the real thing once you put it on a board. It's (at least) the interconnection stuff that makes it non-ideal.

[trad3mark]

Cool. I'm starting to understand the circuit theory a bit more anyway. I'll give it a go during the week and let you know how I get on.

Cheers again R.G.