Questions for experienced Fuzz Face builders

[midwayfair]

I'm willing to bet my bottom dollar that it's a biasing issue.

No, what he's describing is the type of oscillation that you only get by connecting the input to the output. I suspect it has something to do with the ground, pulldown resistor, and output cap all being essentially in series, but I'm having a hard time wrapping my head around the grounding to be sure. Biasing issues do not typically cause any sort of oscillation.

[Vallhagen]

I am happy to read your post guys:)

In a short hopeful moment after i read your post, midwayfair, i thought it could be solved by removing the input pulldown resistor. So i went back to the breadboard to try it. Result: No luck at all ... same sound.

Anyway... i will finnish this FF with the 1044... but if someone feels like going deeper into this, i will be a more than happy reader.

...cant say it enough times: This forum - you guys - are just great. So much knowhow and sharing:)

Cheers!

[midwayfair]

In a short hopeful moment after i read your post, midwayfair, i thought it could be solved by removing the input pulldown resistor. So i went back to the breadboard to try it. Result: No luck at all ... same sound.

Oh, bugger. Oh well. Like I said, the isolated multiple grounds is not something my brain will process without some hands-on with a multimeter in tow.

[Yonatan]

Result: It oscillates in many settings and it sounds... a little different, in a bad way. Hard to explain how, but i should say more "broken" and thrashy sound. Maybe the word "harsh" is correct. I guess the bad sound is caused by added oscillations too.

And why this happens is beyond my knowledge.  i think - as you say - that it looks just fine.

Hi, it's my first post here (I just got my first order of a couple of madbeans boards and then registered here). 

About the oscillations, I'm wondering if you ended up wiring this as positive ground PNP.  I'm working on my first fuzz face type circuit, and I experienced Horrible oscillations (they happened at different settings, and they also happened when I turnied the tone control on the pickup all the way down, and the whole thing sounded harsh).  After reading this http://www.muzique.com/lab/fuzzface.htm I added a 220uF power filter cap from V+ to ground and the osciallations went away!  A similar problem crept in once more after that (loud squre wave synth like sound when I turned my pickup volume all the way down), but this time I tightened some connections on the breadboard and it went away.

[Vallhagen]

Hi, it's my first post here (I just got my first order of a couple of madbeans boards and then registered here). 

About the oscillations, I'm wondering if you ended up wiring this as positive ground PNP. 

Hi Yonatan.

Thanx for your link and hints. I havent read that link before. But the schematic in the link looks very much the same as the one i tried (a few posts up^^). No surprise, i "borrowed" the idea from Tonepad, which in turn borrowed it from Jack Orman, i guess:)

No, i gave up the "upside down" idea for PNP-s, i finnished it with a voltage converter for the negative ground. You can find my finnished build under the "Build Reports" board .. In my circuit i have 47uF as 9V filter cap, which didnt help me with the oscillations. Maybe a larger one would have solved it, but i didnt go all that way:)

...and your description; oscillating and harsh; also dependent of your tone (and volume) guitar pot, sounds pretty much exact like my experience.


Cheers

[Jimihendrix1987]

Hi there folks,

I'm into a new fuzz face fx and I was wondering why anybody has never concerned about the voltages in Q1 and then I read this topic.

What i'm really doing is trying to improve the sound of a JHF1 Fuzz Face, what I guess It won't be difficult cause it really sounds like crap. This version mounts a pair or BC108 with hfe of 228 (Q1) and 238 (Q2), and I really ignore the leakage even though it won't be high just because they are silicon trannies...

I have used this applicattion http://www.diystompboxes.com/analogalchemy/emh/emh.html proposed by Marauder, however I still don't get the voltages required in Q1 that are a must have...

According to geofex schematics, I'm using a 100k resistor as R4(feedback resistor) and the calculator says I ought to use:
R1 = 12.3k
R2 = 4.7k
R3 = 220ohms (Actually I have soldered a 820ohm resistor to have more output)

Tweaking the values a bit I get (with a 8.98v battery)
Q1:
e = 8.93v
c = 7.35v
b = 8.33v

Q2:
e = 7.98v
c = 4.58v
b = 7.36v

You will notice that the only voltage value that it is considered acceptable is the colector at Q2. What should i tweak and mod to get acceptable voltages?

By acceptable voltages in a silicon FF I mean:
Q1:
C 1.4V
B 0.6V
E 0.0V

Q2:
C 4.5V
B 1.4V
E 0.8V

HELP!

[9Lives]

I think you should ignore the calculator for now. Try the original values. R1: 33 K  r2 :330 r  r3: 8 K 2 r4: 100k
First I would try a 10k trim pot for r3 and get your collector voltage down. It diode suppose to be 4.5 I think. I had
A Hard time using that calculator too. Give out ashot

[9Lives]

Silicon fuzz face.  Q1 c: Q1.23. B: .5(ish) e: 0.   Q2  c: 4.7 -4.5. B: 1.23. E: .6. 
See, your voltages are to high. I only use he ge transistors for these but Iimagine
A lower hfe transistor for q1 would help, but some somewhere you have something wrong.

[Vallhagen]

Tweaking the values a bit I get (with a 8.98v battery)
Q1:
e = 8.93v
c = 7.35v
b = 8.33v

Q2:
e = 7.98v
c = 4.58v
b = 7.36v
[...]

Q1:
C 1.4V
B 0.6V
E 0.0V

Q2:
C 4.5V
B 1.4V
E 0.8V

I think you have measuered your voltage with the wrong reference point. Take a look at this numbers, makes sense:)

Q1:
e = 8.98-8.93=0.05V
c = 8.98-7.35=1.63V
b = 8.98-8.33=0.65V

Q2:
e = 8.98-7.98=1.00V
c = 8.98-4.58=4.40V
b = 8.98-7.36=1.62V

Cheers

[Jimihendrix1987]

Yeah Vallhagen !

Swedish people are smart What a brain, discovering arithmetical operations in a glance ! It makes all the sense and the fx sounds awesome !

How do you usually measure voltages in the Fuzz Face? (I've done between +9v and straight to every part of the transistor I wanted to know)

Cheers,

[Yonatan]

Vallhagen: I just had some oscillations again after switching transistors, and it was solved by biasing Q1 (I put a 100k pot in place of the 33k and tuned it until the oscillations stopped), so you might also want to check that.

[Vallhagen]

Yeah Vallhagen !

Swedish people are smart What a brain, discovering arithmetical operations in a glance !

Hey! I have to keep that as a quote for my signature

To put it simple; your zero-reference is always (one of) your ground point(s). Most common, when gnd is negative, your black battery wire is gnd.

**************

Yonatan. I have actually thought of that in general; why is there (almost) always a trimmer for Q2, but not for Q1, in the FF designs out there...

Anyway, good hint to know that you succeded to trim away the oscillations.

Cheers

[midwayfair]

Yonatan. I have actually thought of that in general; why is there (almost) always a trimmer for Q2, but not for Q1, in the FF designs out there...

Because for the most part the voltage of Q1 will not vary by that much, and you get very little sound change by fiddling with them. Really, all you need to care about is whether the guitar input can saturate it, which can happen with anything less than a volt on the collector. You're not going to see anything outside the .5-.7v range on the collector of Q1 with the stock values with any transistor you're likely to use, and anything in this range or within probably a volt or two upward will be sufficient to produce fuzz. Remember, the 100K feedback resistor prevents Q1 from hard clipping (basically, it prevents it from turning off all the way), and the bias on the second transistor therefore affects the gain of the first circuit.

The purpose of biasing Q2 is to balance between actually having it distort and having it hard clip, which produces the "transistor sound" we all know and hate (well, many of us hate it at any rate). Q1 doesn't typically have this problem.

There's a catch, however. The total gain of Q1 does affect Q2 because it overdrives the second transistor. So even though biasing Q1 is largely worthless, the gain rating of Q1 *is* very important. Too much gain and it slams Q2 so hard that the circuit starts oscillating, which is the problem everyone's having with the silicon circuits.

There are two common and easy solutions to this: cut some high frequencies to prevent them from hitting the base of Q2 as hard (that's what that little cap is doing in the Hipster), or use a lower gain transistor* in Q1. There are a few other things you can do ... you can block more of the signal from Q1 by replacing the collector > base jumper with a small resistor, like 1K, or change the feedback resistor value. But any additional changes you make are going to have basically the same effect as lowering the gain of Q1 or using the treble cut cap: darker tone. In other words, you might as well stick to the easy solutions if you want it to behave like a Fuzz Face.

*Edit: You'd be shocked at how much lower the gain needs to be when it's silicon to get it to behave like a germanium in the 60-90 range. Go for something more like 25-60.

[Vallhagen]

Thanx Jon. Now i know a bit more, again:)