PNP Fuzz Face w/ Negative Ground - Issue

[Pmelius]

Currently breadboarding this ala http://www.muzique.com/lab/fuzzface.htm

Works and sounds great, but only if I have the input jack ground referenced to V+ instead of V- and the input coupling cap in the reverse polarity of what's shown on the schematic.  (So, mine has + facing the jack and - at the base of Q1.)

If I reverse the polarity of that cap and ground my input jack to V-, the effect "works" but oscillates like crazy. 

Measured connected to a one spot power supply.

Q1 - 2n1184
C 6.87
B 8.68
E 8.67

Q2 - 2n5087
C 1.98
B 8.67
E 7.51

[iainpunk]

Hey,

I recognize the problem and had a similar one with another fuzz pedal. I remedied it by putting a 33uf capacitor between the V+ and ground.
Bringing back the pedal to positive ground is a more popular solution though. This can be achieved with a simple charge pump.

Good luck,
Iain

[Pmelius]

Thanks for the reply!

I actually do have a 47uf electrolytic filtering the power supply currently, along with a 1n4007 diode for protection. I should have mentioned that. 

Besides that, the only changes are resistors values on the collectors.

I should try one that way, just don't have the parts handy, unfortunately!

[thermionix]

but only if I have the input jack ground referenced to V+ instead of V- and the input coupling cap in the reverse polarity of what's shown on the schematic.  (So, mine has + facing the jack and - at the base of Q1.)

That is the normal configuration for a positive ground PNP Fuzz Face, just like the first diagram in the link you provided.

[pinkjimiphoton]

try a HUGE cap across the rails. 1000-10,000uF.
only way i;ve gotten some builds to stop oscillating.
particularly if you turn your guitar volume down all the way and get a loud tone you can change with the pot before the guitar kicks back in.

i am not a fan of this positive ground trick. i find it to be more trouble than its worth, and it never seems to work all that well.

good luck mate

[Pmelius]

try a HUGE cap across the rails. 1000-10,000uF.
only way i;ve gotten some builds to stop oscillating.
particularly if you turn your guitar volume down all the way and get a loud tone you can change with the pot before the guitar kicks back in.

i am not a fan of this positive ground trick. i find it to be more trouble than its worth, and it never seems to work all that well.

good luck mate

I tamed it a bit with a 2200uf. Tried two in parallel with little difference. The 2200uf allows the circuit to work unless my volume is all the way down, I got some low freq oscillation. Think I'm gonna box it like this and order some damn charge pumps and NPNs so I never have to do this again. 

[thermionix]

I thought you had said positive ground in the title, or did you edit it?  Maybe I misread.  Either way, my comment above was with PNP/positive ground in mind.

[R.G.]

I've posted this advice here many times. I guess one more won't hurt.

Simply calling "-9V" on a PNP pedal "ground" and making the original "ground" be +9V doesn't always work, and it doesn't work in ways that are not simple to fix. Well, not simple other than putting it back to ground and -9V, anyway.

We get posts as frequently as once per week to every couple of months saying pretty much what you're saying: "I found this schematic [somewhere] and it's a PNP fuzz with a negative ground, cleverly just moving the power supply to where I'd like it to be so I can run all my pedals as negative ground, and it now [insert bad actions here, usually excess noise and/or oscillation]. Yet people don't do the research, and run into it over and over and over and over.

Bottom line: doing this reversed grounding of a PNP based pedal works often enough to keep people doing it, and doesn't work enough to mystify beginners. Sometimes it can be cured by better power supply filtering, hence the 33uF, 47uF, up to 1000uF that were suggested. Maybe that works. Maybe not. Maybe it's a partial fix.

People who get away with it working will swear up and down that it always works, because to them, it did. Until they happen to get one that doesn't.

The reasons that it doesn't always work are subtle. I'm not a beginner, and I've beaten this nearly to death on my workbench. I have not been able to come up with a slam-dunk cure. I've beaten on it with theory (and I do have the theoretical background to try this) and by experiment (and I do have the experimental background and experience to do this) and there are some cases that I can't cure.

Best advice: don't do the equivalent of running with scissors, just like in early grade school. Mostly you get away with it, but sometimes, you hurt yourself. Put the power supplies back like the should be. It'll work that way, given that there are not other problems.

Maybe I should save this post off for the next person to write in about this next week.

[Pmelius]

I thought you had said positive ground in the title, or did you edit it?  Maybe I misread.  Either way, my comment above was with PNP/positive ground in mind.

Yes, after your post I realized my title was incorrect and changed it.

I've posted this advice here many times. I guess one more won't hurt.

Simply calling "-9V" on a PNP pedal "ground" and making the original "ground" be +9V doesn't always work, and it doesn't work in ways that are not simple to fix. Well, not simple other than putting it back to ground and -9V, anyway.

We get posts as frequently as once per week to every couple of months saying pretty much what you're saying: "I found this schematic [somewhere] and it's a PNP fuzz with a negative ground, cleverly just moving the power supply to where I'd like it to be so I can run all my pedals as negative ground, and it now [insert bad actions here, usually excess noise and/or oscillation]. Yet people don't do the research, and run into it over and over and over and over.

Bottom line: doing this reversed grounding of a PNP based pedal works often enough to keep people doing it, and doesn't work enough to mystify beginners. Sometimes it can be cured by better power supply filtering, hence the 33uF, 47uF, up to 1000uF that were suggested. Maybe that works. Maybe not. Maybe it's a partial fix.

People who get away with it working will swear up and down that it always works, because to them, it did. Until they happen to get one that doesn't.

The reasons that it doesn't always work are subtle. I'm not a beginner, and I've beaten this nearly to death on my workbench. I have not been able to come up with a slam-dunk cure. I've beaten on it with theory (and I do have the theoretical background to try this) and by experiment (and I do have the experimental background and experience to do this) and there are some cases that I can't cure.

Best advice: don't do the equivalent of running with scissors, just like in early grade school. Mostly you get away with it, but sometimes, you hurt yourself. Put the power supplies back like the should be. It'll work that way, given that there are not other problems.

Maybe I should save this post off for the next person to write in about this next week.

Great advice! Maybe it should be a sticky haha.  Yeah, I don't intend to attempt this again.  My remaining PNP's I'll find another use for or alternatively build them in a standard format with a battery snap. 

For anyone else who this might benefit, I shot an email to Jack Orman, the host of the site where I found the schematic, and he had this to say:

Good power connections are the key.

The side lug on the power jack is the central power point. You should run a wire to this point from Q1 emitter, the Drive pot ground and Drive pot bypass cap - all separate wires. Put a ceramic 0.1uF cap from the power plus to minus. A low ESR cap is best.

The ground point of the 470 and 33k junction should go direct to the power jack ground lug.

This is not a guaranteed fix but it is a start. A lower gain transistor for Q2 is a help too.

[R.G.]

Yeah, it works out that way. Something about the inverted ground setup (which is my personal name for it) makes high gain amps very, very sensitive to oscillation. Very often these circuits have a lot of what sounds like extra-angry hiss. That's often caused by the circuit oscillating well up into the RF region, and then the transistors convert this back down to audio by acting as inadvertent radio detectors. The hiss can be thought of as the phase jitter on the RF oscillation. People with long experience with RF construction could perhaps do a better job, but not many of those come here to either consume or dispense advice.

There is no question in my mind that such an inverted ground setup could be set up to work fine. But it's something that has little bearing to people who are just trying to get an effect to work. These people (maybe you) don't really have the background in RF construction and theory to give it a good chance. Special wiring, ceramic decoupling caps, wire routing, shielding, star grounding, even stage isolation shields and ferrite beads were just some of the things I tried when I was contending with my own personal inverted ground disaster. Lots of wear and tear on the human in that process!! I decided that it was a lot of work to pay for not having to use a charge pump voltage inverter or a second $25 power supply to run all of the possible PNP negative ground pedals in a setup.

So it works often enough for the internet rumor mill to keep postings alive that say "just do this, it's GREAT!". And then doesn't work often enough to keep it a continuous source of posts here wondering what went wrong.

Sigh.