Strange problem biasing a Fuzz Face

[gigimarga]

Hello,

I am a big fan of fuzz face...i've built over 20-30 over the time (most of them for me... )...but yesterday a strange thing happened...i made one FF using two AC128 almost new...first with the gain of 137, second with the gain of 189 (measured with a DMM...i didn't measure the leakage of the the transistors because i did that when i've bought them and there was under 100uA for all)

Using the Analog Alchemy's EMH for Fuzz Face (which i think that's very useful...) i've put a 18K resistor from -9V to the collector of Q1 and a 10K trimpot from -9V to the collector of Q2 to bias it at -4.5V (others values are classical: 100K from Q2's emitter to Q1's base, 470 ohms between the bias resistors, etc.).

But when i tried to bias Q2, using a 9.2V battery, i saw that when the trimpot it's maximized (10K) i have 6.6V on the Q2's collector!!!

It's for the first time when a 10K trimpot can't bias the Q2 of a Fuzz Face...it's that possible or i made a mistake (hard to belive...but possible)?

After that i changed the trimpot to 50K and i biased cery easy Q2 using something like 15-20K from the trimpot and seems to sound OK...

Thx a lot!

[petemoore]

  I used to go with a 10k pot + 6k8, or 10k pot and 4k7, that would allow bias resistance range that'd bias 'most transistors.
  Using the stop resistor made the useful range of the potshaft wider, and easier to fine tune the collector voltage.

[Dragonfly]

I never have any luck using any of the "calculators"...dunno why.

I would go with what Pete said...I usually use the 4k7 + 10k trimmer setup

[gigimarga]

Thx a lot both...i will try to use a 4.7K in series...i think it will be enough

[tiges_ tendres]

Are you sure of the value of the trim pot?  I have banged my head against the wall a few times when all I needed to do was check the value of the pot with a meter rather than with my poor eyes!

Dont just check the bag the trim pot came out of either!  That one caught me once too.

[mac]

As a general rule, more leakage at Q1 means higher collector resistor at Q2. With a 200ua you may need 20k or so depending on Vbe and hfe.

I never have any luck using any of the "calculators"...dunno why.

As I wrote one of them I can say that the problem with calculators is that there are many variables and sensible eqs, and small errors generate large deviations. Also the assumption that real collector current is ic* = hfe*ic + leakage, and that Vbe is fixed contribute too. And the knowledge of Ge hfe and leakage at the moment.
But they can be used to expect resistors values and voltages before building or breadboarding.
For example, they predict lower collector resistors for Si, and the behaviour of Q2 collector voltage when you tweak Q1 hfe (like finger heating only Q1), and viceversa. Also that you can swap a Si for a Ge at Q2 with little impact on the bias network.

mac

[hellwood]

its happened to me. now i always use a 50k trimpot.

[Solidhex]

Yo

  I enjoy messing around with Mac's program. I think it shows you a lot about what affects what bias wise in the ff circuit. If you mess with the leakage values you can see how they can really mess with the way the circuit biases up.
  You really just have to breadboard the thing and tweak it there or better in the circuit itself.

--Brad

[gutsofgold]

Is bias personal preference or is more of a wrong/right value? On my fuzz I have a 4.7k followed by a 10k trim and with the trim set to 0.0 I can only get 3.7V on Q2 collector. It sounds pretty good but hey, I might like it even more closer to whatever it is you guys are striving for. Can i change the 4.7k without worry?

[gutsofgold]

Welp, I answered my own question. Too much voltage is thick and too little is very thin. Yup, sounded the best at 4.5V !

Andy I'll shoot you an email with the values. hopefully you have those strange resistors in close values.

[gigimarga]

thx all...now i'am peaceful...and in the future i will use a 4.7k in series with the 10k trimpot