Germanium Fuzz Faces suck to build

MBFX

Well-known member
Seriously. Every single build I get a new, weird problem. Tonight I have a fantastic sounding circuit that is whisper quiet. WHY ARE THESE SO FINICKY?
 
If it's not on breadboard and you've soldered it then it could be a cold solder joint. I had this issue once and reflowing the joints around the output fixed it and I had normal volume again.
 
Have you played with the values of voltage divider located between Q2 and V-? I usually replace the 470R and 8k2 with a 1k2 and a 7k5, respectively. It doesn’t change the Q2 collector voltage, but it makes the output level much higher, which is how I like that circuit to sound.
 
I find them finicky as well but usually it's not the volume. It's the bass content - easy fix to reduce the input cap to 470k - and some weird behavior like some of them overpower my amp and some don't.

However I almost always bump up the 470R to 1k or more to get more output. Maybe I shouldn't. With a tube amp set properly, the amp caps the volume anyway. And I won't play a primitive fuzz into a solid state amp. Life's too short for that hot garbage :)
 
It's on breadboard, all the voltages are good, it sounds incredible... it's just fucking quiet. I hate these things. It would be fine if they sounded shitty, but they don't. They're incredible. UGH.

I like this schematic because it includes bias voltages:
1706904489155.jpeg

I omit the fuzz pot for a 1K resistor, because who actually runs a Fuzz Face at less than max gain? That's what your volume knob is for. I use a 100K output pot, but that isn't the problem - the circuit is a hair below unity gain even coming straight from the output. Replacing R2 with 1K didn't seem to do anything, but I wasn't expecting it to. I use a trimpot in place of R3.
 
It's on breadboard, all the voltages are good, it sounds incredible... it's just fucking quiet. I hate these things. It would be fine if they sounded shitty, but they don't. They're incredible. UGH.

I like this schematic because it includes bias voltages:
View attachment 67460

I omit the fuzz pot for a 1K resistor, because who actually runs a Fuzz Face at less than max gain? That's what your volume knob is for. I use a 100K output pot, but that isn't the problem - the circuit is a hair below unity gain even coming straight from the output. Replacing R2 with 1K didn't seem to do anything, but I wasn't expecting it to. I use a trimpot in place of R3.
I always run a fuzz face at less than full fuzz. It's the only way to have it clean up. Really clean up.

Raising R2 to 1k or more will absolutely give you more output because you will have to reduce R3 accordingly, which is what actually raises the output. Try connecting the output cap directly to Q2c. It should be loud as all hell. Unbeatably loud. If it isn't, something else is wrong.
 
The suck part is that you get this be-achhhhh to sound good on the bench and then go to play it live and it blows. Get used to the fact that these things are the luck of the draw and deal with it. In my case I love compression and the difference between how a compressor deals with your amp on a given day at a given place seems to be a fleeting moment. Fuck it. Play your best. The audience will never know the difference if you have a smile on your face and sell the part.

I realize this is a massively over arching solution but how else do you deal with it?
 
And yes, the toughest thing I've ever built is a fuzz face and I don't think it depends on the build as much as the player.
 
I rebuilt the circuit on the other half of my breadboard again, and it worked after some messing with it. I tried all my transistors by building just the first half of the circuit, and they all work and give the expected amount of boost. In the FF circuit, some of them just don't work - looking at you, AC128s. They were sold as low leakage, but I haven't made a test rig yet and my TC1 never recognizes germanium transistors. I assume it is a leakage problem. Using GT402s with high gain and low leakage is a much better time.

One thing that is puzzling me, though, is that all my voltages are good on the AC128s except the Q1-B voltage. It comes in around -0.045VDC. If I run a very hot signal into it, we get fuzz. What am I missing?
 
Not doubting as I’ve seen germanium’s all over the map as I’ve done more builds. Fortunately I was lucky when I built my first Fuzz Face like 25 yrs ago.. I traded someone some used parts for a pair of germanium’s he had. Not sure what they are but I’ve never heard a FF come even close to this one! Even rebuilt it on a good board and recycled them and it still sounds amazing.
Sadly I am liking silicon fuzz faces more these days.
 
I rebuilt the circuit on the other half of my breadboard again, and it worked after some messing with it. I tried all my transistors by building just the first half of the circuit, and they all work and give the expected amount of boost. In the FF circuit, some of them just don't work - looking at you, AC128s. They were sold as low leakage, but I haven't made a test rig yet and my TC1 never recognizes germanium transistors. I assume it is a leakage problem. Using GT402s with high gain and low leakage is a much better time.

One thing that is puzzling me, though, is that all my voltages are good on the AC128s except the Q1-B voltage. It comes in around -0.045VDC. If I run a very hot signal into it, we get fuzz. What am I missing?
Sounds like bum trannies - ahem, faulty transistors. I've had good success with matched pairs bought on eBay from the usual Ukrainian and Bulgarian suspects.
Germaniums are finicky but the only circuits I've had serious issues with are the infamous MKI, MKII and MKII Tone Benders. Those always have issues for me.
 
Back
Top