PNP Fuzz Face gone splatty - bad transistors?

[GrooveChampion]

Well, I build the one with the built in voltage inverter, worked fine for a while the it got all splatty, bias knob barely does anything. I used AC128 trannies

Anyone had a similar experience?

Anyway, for some reason I had to replace the 10K bias pot with a 100K one to get the collector of Q2 to 4.5 volts(-4.5) and there quite the volume drop even at full volume.

This is the layout:

http://tagboardeffects.blogspot.com/2010/11/fuzz-face-pnp-negative-ground-with.html

[Travis]

Are the transistors socketed? If so the Ge transistors tend to come loose from sockets due to their mass (and the poor quality of the ubiquitous Tayda SIP socket)

[GrooveChampion]

Well, I actually soldered them to the SIP sockets after I was sure everything sounded right the first time. I have no idea what could go wrong with the solder between the time it sounded right and now that it doesnt.

[Travis]

Soldered every pin or one per transistor so they don’t fall out?

Otherwise Fuzz Faces sound like shit when they’re hot. Is it hot?

[GrooveChampion]

Well yeah, around here its quite hot

Could it really affect the tone so much as to sound like a velcro fuzz?

[Travis]

Leakage doubles with every 10 degree increase. So if it’s even just 10 degrees hotter than when you built the pedal the leakage may make the sound unusable (high leakage makes a fuzz face increasingly gated/weak)

[GrooveChampion]

Is there a way to deal with this issue? Besides throwing the damned thing in the freezer? Cos it has been hot here lately?

I mean I dont hear a lot of people complaining about such issues.

[Travis]

Maybe first throw it in the fridge or freezer to assess if that is actually the issue. At that point you’ll have a better idea what action to take

If it is the temperature instability, then I recommend you use transistors with the lowest leakage possible for the best stability. Otherwise that is really the classic problem with the Ge fuzz face, and the answer was the Si fuzz face. A hybrid can have better temp stability than an all Ge pedal with a smoother sound than all Si

[SpaceCommandant]

From Fuzzcentral:
“The biggest problem with the Germanium transistor version of the Fuzz Face circuit is that Germanium transistors aren't temperature stable like Silicon transistors. When the inside of the enclosure starts to heat up, whether it's from stage light or sunlight, the pedal will start to sound very bad, and eventually if it continues to heat up, it will stop working all together. This is called "thermal runaway." The best way to stop this problem is to add a reverse- biased Germanium diode across the base-emitter of Q1. This process is described in detail in the Britface article at the Plate-to-Plate effects site. Make a note that this problem only happens with Germanium transistors, and in order for this mod to work, the transistor AND diode have to be Germanium and the diode HAS to be reverse biased.. For a PNP circuit, the cathode of the diode would be connected to ground instead of the anode.”

[Travis]

Good one Space how could I forget ye olde MKII trick?!

[GrooveChampion]

OK I know its been a while but I wanna ask it anyway, how are so many pedal makers keep their germanium based pedals from reacting so badly to heat? I mean there are many circuit and pedals with germanium transistors that I dont hear anyone complaining about? Is the trick to use low leakage transistors?

[BetterOffShred]

Yeah using hand picked low leakage stuff or designing the circuit to operate within expected leakage etc..  I definitely want to try the diode reverse biased trick .. that sounds neat

[GrooveChampion]

I still dont understand how the diode trick works, I mean I understand it has to be germanium but adding it accross the emitter and the base wont cause them to short out?