Fuzz face buzzing (even in bypass)

[matt.dyck.music]

So basically, I got the board all soldered up and everything working just the way I wanted. I put it in the enclosure and it was fine. But then I added in my LED and all of a sudden everything is buzzing. I tried removing all that I added for the LED, so it would be back to the same but it still just buzzes. In bypass it has a normal guitar signal with a lot of extra noise, and through the circuit it's just a loud and aggressive humming.

Anyway, I'm going to be away over the holidays so I won't be able to try anything til I'm back (which will drive me nuts thinking about, lol). So what I'm mainly looking for is tips/tricks of figuring out what's wrong. If it's buzzing in true bypass then I would assume that power is going straight to ground (or would it be to the input or output somewhere?). I'm really new to this, so I haven't yet built up the knowledge of how to diagnose these things. Thanks in advance.

[matt.dyck.music]

I also just realized I did the classic "post in the wrong area" thing. Is there a way I can get a mod to move this to the appropriate forum?

[bluebunny]

You could PM Aron and tell him what a fab guy he is, and ask him really, really nicely if he can move it? 

Or you could just break all the cross-posting rules (coz there's so little traffic here) and repost in the main section?

[matt.dyck.music]

Thanks, I'll do that. Also, I actually had just a little bit of time to try figuring out of this morning. So, I built an audio probe, cause I figured it'll be useful in the future anyway. With it there is loud buzzing as soon as power is connected to the pedal before probing anything (just the ground sleeve connected to the pedal ground.)

[idy]

Possibly a ground fault? Are the jacks grounded together? All grounds good to one place?

[matt.dyck.music]

Yep, all grounds are connected together. And the first thing I did was to go through and trace each path to make sure none of the connections came loose or anything (all the solder joints looked good. Is it possible that a damaged transistor would cause this? I didn't know I could get sockets so I just soldered them in (and they were working fine). But the last thing I did before the circuit stopped working (pedal sounded fantastic) was to solder the LED in, which I grounded to the same joint as the transistors. The only other thing I did was to put it in the enclosure again to test it with the LED (with some plastic to protect anything from grounding to the enclosure). I should also note that the LED barely lit up (but had been bright enough when connected directly to 9v battery)

[idy]

voltage on the transistors? Audio reaching first transistor?

[matt.dyck.music]

I'll have to check when I'm back in a week. Not sure how I test if audio is reaching first transitor. I would assume with the audio probe, there's just so much constant buzzing it's hard to test (trying not to be too noisy for the upstairs neighbors, and i was working on it late at night).

[tubegeek]

When you say "buzzing," is it like a steady intense tone (oscillation) or like an unshielded cable (buzzy noise) - I ask because usually buzzy noise is lower in volume than the signal by a fair amount whereas an oscillation could be louder than the signal - neighbor-annoyingly loud.

When you get a chance, short the input to ground and report on that, both in "effect on" and in "bypass."

[matt.dyck.music]

I believe it was more like the oscillation you're describing. It was certainly louder than the buzz you get from touching the end of the cable unplugged. (probably at least double that).

[tubegeek]

Try connecting a large-value (47uF or more) electrolytic cap between your 0V and -9V points on the pedal (assuming it's the standard Fuzz Face positive ground circuit) keeping the wiring as short as is practical. Observe the correct polarity of the electrolytic, with the (-) stripe toward the more negative voltage (-9). Also see if there are any places where your layout can be tidied up to make sure the input wiring is not thisclose to the output, nor any wiring is especially long.

If you are using a negative-ground scheme, some Fuzz Faces do, do the same thing as mentioned above but your connections are +9 and 0, with the stripe towards the 0.

For a slightly more thorough power supply filtering job, it would be even better to do this: add the cap described above. Before feeding the filtered DC to the circuit, put a small-valued resistor (47 ohms is probably OK, up to maybe 220 ohms) in series with the DC connection, and another 47uF-or-more cap from positive to negative after the resistor. This configuration is referred to as a "pi" filter because the arrangement on a schematic looks like the greek letter π.

If the circuit is oscillating because of the power supply having a high impedance in it somewhere, this'll most likely kill the oscillation. If it's oscillating because you've made a mistaken positive feedback connection, it won't cure that.

Good luck!

[matt.dyck.music]

Thanks for the reply. I don't know how I didn't even mention it is negative ground yet... Must be the lack of sleep with a new baby, lol. I'll give this a try, though I'm not certain it will make a difference as it was working fine with everything wired the same earlier. I'll still try it anyway, but could you explain what adding this capacitor actually does to the circuit?

[tubegeek]

One way to get the gist of a capacitor's purpose in general (to a 1st approximation) is to think about AC and DC separately at first. A capacitor blocks DC voltages: you can connect two parts of a circuit with different DC voltages via a capacitor and (from a DC perspective) they are not connected at all.
From an AC perspective, a capacitor acts as a short circuit FOR AC FREQUENCIES ABOVE A CERTAIN THRESHOLD. (The threshold depends on cap value and the other impedances around the connection.)

In a deeper understanding, this is not a cut and dried, on/off, open circuit/short circuit threshold: there is a transition as we go from low frequencies (DC is 0 Hz) to higher frequencies, and the capacitor will transition from very high impedance (open circuit) to very low impedance (short circuit.) Again, for an initial understanding, we can just look at the extremes and know (but not worry about too much) that there is a frequency where we switch from so-high-it's-like-an-open-circuit to so-low-it's-like-a-short-circuit somewhere between those two extremes.

OK, now let's look at what a power supply is intended to do: supply noise-free DC voltage to a circuit. DC: good. AC: bad (noise is an AC signal.) So a power supply bypass capacitor has the function of keeping the DC voltages separate, but short-circuiting the noise right back around into the power supply without getting a chance to enter the rest of the circuit.

The series resistor in the power supply gives the circuit a defined non-zero impedance to use when we're making the comparison between "low" and "high" frequencies. The resistor has the same impedance at all frequencies including 0 Hz (DC.) We can then plan the capacitor's transition from high to low impedance intentionally to give us good-enough noise filtering by design.

In a negative-ground Fuzz Face, the power supply filter gives any unwanted oscillation (an AC signal) somewhere easier (lower impedance) to go (back into the power supply) rather than into the higher-impedance audio circuitry where it can be heard at the output as a nasty noise.

In the negative-ground Fuzz Face I *VERY* recently built, I needed to add a pi filter like this to keep it from oscillating. Worked perfectly. So I feel ya!

Congrats on the new baby!

[matt.dyck.music]

Thank you for the explanation, these are the kinds of responses I need (that way I actually can understand more of what's going on in the pedal). Anyway, I just came back to it and it's for some reason working again (at least there's no buzzing in bypass anymore and there's audio signal coming out of the circuit again).

The strange thing is that Q2 is reading 1.3V even though it was like 4.7 before. So now it's just kind of noisy but the overall volume is much less than in bypass.

[matt.dyck.music]

So I added the LED again cause I figured that isn't actually the problem and the light works fine now. I think the problem at the moment is just the incredibly low volume and that the buzzing I hear when the volume is up enough to hear the signal will seem significantly quieter once the pedal is making it's full volume again.

It turns out I don't have any large enough caps to do the "pi" circuit (will have to stock up on more parts, lol). And I used my continuity setting on my multimeter to confirm that the transistors are working and not shorting out.

I'm still not certain why Q2's voltage changed, but I will try to figure it out.

[tubegeek]

If I had to guess I'd say you have a bad solder joint(s) somewhere: stuff that works sometimes/sometimes doesn't is pretty much always bad solder joints.

What's your level of experience with soldering?

[matt.dyck.music]

I have a decent amount of experience soldering, it's just usually wires rather than circuit boards. I don't see any harm in going over my solder joints anyway. I did a small amount of that in one area of it yesterday and got my volume back and the transistor read 4.7V again. The only thing was still a fairly noticeable buzz, but I think the "pi filter" mentioned above should solve that. I just have to see if the pedal stays consistent now or if it's still finicky.

[tubegeek]

Progress! Good news!

[matt.dyck.music]

I just realized I never really followed up. Anyway, all I ended up doing was what I already said, but I then put it all together in the enclosure and it's working fantastic! Didn't end up doing the pi filter becaise it actually isn't that noisy once it's in the enclosure. If I have an issue later on I may try it, but for now it's perfect for me. Thank you everyone that helped me out and taught me some things. I look forward to whatever I decide to build next.

[tubegeek]

Thanks for circling back. Success!

My experience (have you ever been experienced?): After building all sorts of stuff for all sorts of years, I finally built my first Fuzz Face a few months back and I love it. So gnarly and warm.

I had to get a bit deeper into power filtering for my Fuzz Face because I built it with a negative ground, but with PNP germaniums. I used this article as my inspiration:

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

And I did indeed get some oscillation at first. The minimal pi filter on the power rail was a robust solution (probably overkill, as you've discovered) that was kind of a no-brainer.