Fuzz bloom

[PBE6]

I built a fuzz pedal for a friend based on the following schematic:



which I assume is based on the ZVex Woolly Mammoth and the ZVex Mastotron. He likes the sound, but is complaining that at some settings the sound level decreases significantly when he strums hard, and blooms back to full volume a while later. I tried it and this is indeed the case for hot signals when the feedback knob ("POT2") is turned all the way up (max resistance).

I checked out the GeoFEX Technology of the Fuzz Face article again, but I'm still confused as to why it's behaving this way. Is it because the increased resistance is effectively decoupling the transistors rom each other on the feedback loop and doing something funky to the bias? I checked the output with the scope, and the increased resistance squishes the waveform horizontally so there is no signal for a greater part of the cycle.

I tried putting some limiting diodes to ground on the input, and it's better, but I have to use 1N34As to make a significant dent in the problem, and that's limiting the input signal to less than 0.4 V.

I'm wondering is whether this is a voltage issue or a current issue on the front end, or if the fuzz just starts to break down with too much resistance.

[PRR]

> the sound level decreases significantly when he strums hard

So don't strum so hard. (Or use POT1 to decrease the level slamming in.)

> the sound level decreases significantly when he strums hard, and blooms back to full volume a while later.

In a tube amp, that's "grid blocking".

No grid here, but it is really about *any* circuit with a capacitor and diode. My guess is that when Q1 Base-Emitter diode is OVER-OVER-driven, C1 charges more one way than the other, charges-up in a way that forces Q1 to way-low current, poor gain. When you stop hitting it, the excess charge slowly leaks off, until Q1 is happy again.

Put your volt-meter on Q1 Base, also on Q2 Collector, and watch the DC levels change when you BEAT on it.

POT1 really is an effective treatment for grid-blocking. Keep POT1 resistance at least half of R3+POT2 resistance and grid-blocking hardly happens.

Other than that, I think grid-block-like effects are intrinsic to any single-ended cap-coupled amplifier when OVER-OVER-driven.

More than one long-career guitarist teases the edge of grid-block as a style.

[aron]

Wow, I made a pedal that did this and someone said "people will pay a lot of money for a pedal that does that" hehehe

[Steben]

Exactly, it is in a way an exagerated kind of sag.
Find a way to control it, and you have a very useable effect.
It looks like P1 does the job at the expense of gain of course.
Doesn't your friend use P1 often?
And what about using the guitar volume pot to push it in and out the "bloom" area??

[PBE6]

Thanks Paul, I will check the DC levels on Q1 for the education.

I removed POT1 on this build to simplify things, looks like I made it a bit too simple! Time to drill a new hole...

[Mark Hammer]

"Bloom" can be a wonderful thing. I think the biggest challenge with it is that you can't exactly predict and control how quickly or when the blossom opens up.  And if it doesn't happen when you want, it feels like the pedal is fighting you.

So, a question to you, Paul:  If, as you suggest, the input cap "overcharging" has something to do with this, is there any theoretical reason to employ several caps in parallel at that station, or perhaps caps of a given type?  I don't want to start a snake-oil internet meme here, but "bloom", as I say, can be an aesthetically pleasing thing, and if there is a way to harness it in relatively predictable fashion, I'd love to know.

[FiveseveN]

I'm thinking that's pretty much what POT1 is for. And since POT2 affects gating/decay, it may also have an influence on percieved "bloom".

[samhay]

> I don't want to start a snake-oil internet meme here, but "bloom", as I say, can be an aesthetically pleasing thing, and if there is a way to harness it in relatively predictable fashion, I'd love to know.

If you use a more conventional envelop detector (i.e. with attack and release control) that then sets the clipping level then it is quite easy to create predictable 'bloom'. And yes, it can be quite delightful.

[SISKO]

See if hes using any buffered pedal before the fuzz. A buffer in front of a FF makes the FF bloom

[Electron Tornado]

> the sound level decreases significantly when he strums hard

So don't strum so hard. (Or use POT1 to decrease the level slamming in.)

> the sound level decreases significantly when he strums hard, and blooms back to full volume a while later.

In a tube amp, that's "grid blocking".

No grid here, but it is really about *any* circuit with a capacitor and diode. My guess is that when Q1 Base-Emitter diode is OVER-OVER-driven, C1 charges more one way than the other, charges-up in a way that forces Q1 to way-low current, poor gain. When you stop hitting it, the excess charge slowly leaks off, until Q1 is happy again.

Put your volt-meter on Q1 Base, also on Q2 Collector, and watch the DC levels change when you BEAT on it.

POT1 really is an effective treatment for grid-blocking. Keep POT1 resistance at least half of R3+POT2 resistance and grid-blocking hardly happens.

Other than that, I think grid-block-like effects are intrinsic to any single-ended cap-coupled amplifier when OVER-OVER-driven.

More than one long-career guitarist teases the edge of grid-block as a style.

I wonder if that's what happens with the Heathkit TA-28. Using a small enough cap in the same place as C1, and it stops cutting out at the expense of lower freqs.


http://www.luciferstrip.com/fuzz/heathta28-orig-schematic.jpg

[PRR]

> is there any theoretical reason to employ several caps in parallel at that station, or perhaps caps of a given type?

I don't understand. The cap is just a storage-bucket.

My toilet runs to a septic tank (storage) and then to a drain-field. Gravity gives some 1-way "diode" action.

If I flush 50 times a day, the field absorbs it and the level in the storage tank stays steady.

If I flush 500 times a day, the field can't absorb all that and the level in the storage tank rises until something bad happens.

Changing the storage tank from concrete to fiberglass doesn't change that. Changing to larger storage may delay the bad effects, but if you keep flushing (or strumming) you can fill-up any storage.

While this won't work on my p00p-pipe, in electronics we could use a "thin pipe" from toilet to tank, so the flow into the tank can be limited to what the field can absorb.

That would be a resistor. And now we find that PBE6 omitted that resistor. Dart-board, I would try 50K fixed here. But because every guitarist is different, and POT2 is also part of the system (acts like the drain-field) and can be adjusted over a wide range (more or less drain on the cap), possibly another hole is the best path.

[Mark Hammer]

Fair enough.

[Digital Larry]

> is there any theoretical reason to employ several caps in parallel at that station, or perhaps caps of a given type?

I don't understand. The cap is just a storage-bucket.

My toilet runs to a septic tank (storage) and then to a drain-field. Gravity gives some 1-way "diode" action.

If I flush 50 times a day, the field absorbs it and the level in the storage tank stays steady.

If I flush 500 times a day, the field can't absorb all that and the level in the storage tank rises until something bad happens.

I am fond of the fluid-dynamics analogy for electronics, but this one takes the cake and leaves it there!  It also reminds me to flip my diversion valve.

[PBE6]

Turns out I had installed a 100k trimpot in place of POT1, lucky me! Turned it up to maximum and the bloom is 95% tamed. Thanks all (and especially Paul)!

[LightSoundGeometry]

very interesting. so a Bloom is like a sag or a bloom back to happy gain from being pushed ? love this place.