Specifics on a circuits bias...some need more than others?

[Outlaws]

As I am sure everyone knows by now, I am having some serious trouble with my NPN negetive ground silcone NTE123A Fuzzface.

If this is a bias issue, as two members mentioned, then I guess I have a rudimentary quesiton...

How come a trimmer isn't in the schematic if it is needed?

The sticky at the top of this forum, "what to do when it doesn't work", states that, "Mother Nature does not lie, but she is a stickler for details. If it was put together perfectly, it would work."

So I guess I am asking....why do have a bias issue to start with?
The GGG site this layout came from says to use BC108 silicon and to normally replace the 470 with a 330.  I used the the 330 but used a NTE123A (which I was told by either aron or Dragonfly, I forget offhand, was okay).

I can't believe I need to fix the bias with a trimmer. (no disrespect to petemoore).  The layout has been there for a while so for me to have this issue, it just seems that something isn't working.  And to add a trim pot just seems like a 'work-a-round' rather than solving the actually issue, which again, was stated in the sticky, "If it was put together perfectly, it would work."

I don't know exactly what to ask at this point.  I am at a total loss for words.

Anything anyone wants to add would be very helpful no matter how small.

And I am not trying to come off as rude or anything...I am just honestly at a loss for what I am supposed to do - or forgot to do.  I rebuilt the thing 2 times.  And I found my first problem...the trannies were wired backwards.  I found my second problem...one collector was grounded on accident.

But now this issue with a massive "scream" because I am getting 9v at the tranny rather than 4.5v +/- just seems like something in the circuit isn't doing its job.

What part of this circuit should be doing this?




Anyways, I do want to thank you all for your help so far as I can now read and almost fully understand a schematic and I have literally 100x more knowledge than I did 2 days ago about the real workings of a circuit.

-its just too bad I don't have enough yet to fix this.  :lol:  

[Mike Burgundy]

"perfectly" here means "according to the rules"... And with germanium, the rules are "all devices are different"
This means that every germanium transistor needs somewhat different peripherals to get the same biascondition. Think of it like this:it's a bit like powertubes that need to be matched. It just wasn't possible to produce them with a tight enough tolerance to get this solved. They also leak like heck, which plays havoc with your bias setup.
so.... Maybe that trimmer is a good idea after all.

There's no trimmer in there because of several reasons:
-these values work for most people/situations/trannies
-these values were in the original
-the originals often didn't sound good at all, because of just this bias problem
-resistors are cheaper than trimmers
-a lot of people fine tune with a trimmer, measure that value and then replace it with the nearest resistors
-trimmers tend to drift. You don't want drift in a bias-sensitive circuit.
-actually some people fine tune by replecing the reistor and get it righ in two or three go's.
-and some people don't fine tune at all, because it's not really causing a problem and the sound is good enough.

Some circuits are reputed to need "tweaking". This is one of them. It's part of the electronics, and it's built in discrepancies and variation, so Mother Nature obliges...

That said:
9V at a transistor means it's not conducting AT ALL. This is a bit weird for a germaniumand rather harsh for a bias mis-match although possible - make sure you do some careful continuity tests to see if all joints are ok.

[leonhendrix]

when fuzz faces were first made (germanium and silicon) they didnt worry to much about having to bias transistors using different value resistors or caps because they had large amount of trannies at hand, they would just find a pair that worked with the circuit.

your only using one pair by the sounds of it, so u will probably have to use trimmers to bias it. if you can get different types or a load of the same type and experiment.

there are other fuzz face layouts out there that are more tweakable.

leon

[Pedal love]

Please don't misunderstand, there is nothing wrong with your actions toward this circuit. The problem is these circuits are guideline based, meaning under ideal conditions, this circuit should operate a certain way. That said, it is an imperfect world and many circuits like you are working on have problems with noise as bad or worse than I heard in your audio clip.

Some of the worst noise I have ever heard came from si-ff's I have built. There are ways to compensate for this and the other members here have tried to outline them. Trim bias works often. You could also shunt capacitors or place them between node points, if worse comes to worse. But there is often no simple cure in this game and the best you can do sometimes is just trial and error. Keep trying .....pl

[Outlaws]

Those sound like fair enough explainations.

Though I must state that this is a Silicon FF.

I did try a 20k HighRel trimmer in replacement of the 8k2, but it didn't reduce the voltage to Q2 at all.  I hooked it it up about 3 different ways.  The SparkleBoost I currently have on perfboard...untested soundwise....has a 100k micro pot from Radio Shack that works perfect in sucking the voltage down to 4.5v.  Could 20 just be the wrong value?

But...

9V at a transistor means it's not conducting AT ALL.

What is not conducting at all?  Can I have some bad resistors or bad tranny or bad caps?  Could they have been damaged during the initial problems I had?

I swapped out the 8k2 with a 4.7k and a 10k and a 100k, and nothing in the voltage changed on bit.
I also swapped out first Q1 with a 2N2369A, and then swapped out Q2 with a 2N2369A, but nothing changed anything in the voltage that Q2 recieved.  Q1 got a bit more when it was using the 2N2369A, but it was the same no matter what Q2 was...if I recall.  Q1 was right around 1.48v.

[Mike Burgundy]

Sorry bout the germanium confusion - read over the BC108 bit.
Okay.
For resistors to have any effect on voltage, we also need current to flow, otherwise nothing happens. If the *transistor* is not conducting, then everything between the collector and the battery terminal will sit at 9V no matter what.
There should be some bias current through the transistor.
All of this suggests that either the trannies are bad, put in wrong, or there's somehow a bad connection. Switching transistors suggests they are fine. You are using sockets? If not, stick em in.
Do you have a  DMM with transistor test? Use it.
Make sure the emitter of Q1 has continuity to ground, and the emitter of Q2 has 1k to ground.
You did remember to stick in that wire bridge, right? (sorry, just making sure)

[Outlaws]

Sorry bout the germanium confusion - read over the BC108 bit.
Okay.
For resistors to have any effect on voltage, we also need current to flow, otherwise nothing happens. If the *transistor* is not conducting, then everything between the collector and the battery terminal will sit at 9V no matter what.
There should be some bias current through the transistor.
All of this suggests that either the trannies are bad, put in wrong, or there's somehow a bad connection. Switching transistors suggests they are fine. You are using sockets? If not, stick em in.
Do you have a  DMM with transistor test? Use it.
Make sure the emitter of Q1 has continuity to ground, and the emitter of Q2 has 1k to ground.
You did remember to stick in that wire bridge, right? (sorry, just making sure)

I don't know why the Emitter of Q2 would need a 1k to ground, but I do know this.....



I just got the voltage down to 5.75v!

How you all must be asking yourselves...lol
Well, I will tell you.

I put back in my 8k2 but swapped out the 330 (470 on the layout) with a 1MegOhm.

So now, is there any way to run in series to get it a bit lower?  I tried a 100k after it going to the 8k2, but that didn't seem to do anything.  Though I haven't hooked this up yet for a sound test, it maybe just be lower enough as it is.  God I hope so...

I am really not interested in hooking up my pots again just yet, so I will wait for an answer, but is 5.75 a reasonable level?

[Outlaws]

:cry:


I hooked up my input and output jacks, by fuzz and vilume pots, and fired up my amp.

Nothing.

Then I hit the strings with super human strength and got a slight  sharp distorted note that died right after it came to be.

I test the voltage.

For some reason I am no only getting .5 (or was it .05) volts.

So I took out the 1MegOhm and put back the 330.

I got ZERO volts, but I got a halfway decent sound with a texture of 8 bit distortion that didn't sustain, it just "fizzles" out and dies.

It put in both a 2N2369A and a NTE123A for Q1.  The NTE123A had less of that fizzle sound, but it still had it.


I am now at a point where I am about to scrapp this whole thing.  I am officially clueless.  None of this makes sense.  Its like I keep getting 'right there', but they then something else breaks.

[Mike Burgundy]

'Cuse me?
No pots are in?

Q2 biases, read this carefully, through the 1k gain pot to ground
That's the 1k I'm referring to. Without this in place, there's NO WAY it will bias.
I reckon 1M is waay too high, and doubt the circuit will work but hey...

[Outlaws]

'Cuse me?
No pots are in?

Q2 biases, read this carefully, through the 1k gain pot to ground
That's the 1k I'm referring to. Without this in place, there's NO WAY it will bias.
I reckon 1M is waay too high, and doubt the circuit will work but hey...

Okay.  SOrry about that.  I wasn't thinking.


So let me start this over in a sense...

Everything was hooked up as mention in my previous post (including the 1k gain pot).

So as it stands now, everything like the layout at the top (330 instead of 470 and two NTE123A) I am getting 2.44v at the Q2 collector.

From the battery+ in I am gettin 8.9v.
The other side of the 33k is giving me 1.23v.
The other side of the 330 is giving me 8.67v.
The other side of the 8.2k is giving me 2.44v.   :idea:

So basically this "fizzle" is from the 8.2k resisting too much voltage.

So to let more through, I need to lower the value.

I just breadboarded a 4.7k and it lowered the voltage to 5.04v when the fuzz is at max.

Test time!

Still makes a lot of digital artifact type noise.

But when I turn down the Fuzz it lowers the bias to a little over 4v.

The noise goes away at about 1/3 of the way away from the max position, but I lost most of the fuzz and it becomes very dull sounding at loses most of its tone.

This happens in a very fast manner.  Like I can roll the pot to an area where within 1mm of a turn will turn it on or off.  There is a gradual build up of the noise at that area, but its a lot in a small space.  ANd as said before...I lose all my tone and most of the fuzz.  Plus after rolling off the fuzz even further, I seem to lose volume rather than this acting like a fuzz control.

[Outlaws]

I just swapped out both tannies for the two 2N2369A that I got yesterday.

For some reason now the max voltage is anywhere from 3.8 - 4.02 or there abouts when testing the Q2 collector.

There is a lot, and I mean a lot of noise still.  But I noticed that when I roll off the tone on my guitar, I get a shreek sound.

So now this is happening with to much bias and two little bias.

I am beginning to doubt that atcually getting to 4.5 is going to be any better.


Has anyone else used a LesPaul and a silicon Fuzzface?

Oh, and I am getting radio waves also.

[Mike Burgundy]

okay, don't touch the old trannies. Leave them out until you have a chance to test them in a meter.
Are you using sockets? If not, do.
The Q's you have now clearly work. It now reeks of, I'm sorry if this has already been dealt with, bad solder joints. This would also be a possible reason for the demise of the other transistors.

[Outlaws]

okay, don't touch the old trannies. Leave them out until you have a chance to test them in a meter.
Are you using sockets? If not, do.
The Q's you have now clearly work. It now reeks of, I'm sorry if this has already been dealt with, bad solder joints. This would also be a possible reason for the demise of the other transistors.

?

This is all on a breadboard.

[Mike Burgundy]

ooh. I didn't get that - did you mention that?
Okay.
In that case it's perhaps a little easier (rules out thermal damage and bad joints)
suspect in order of suspicion:
transistor orientation. Trust me, this really sucks. Theyre all different, and often differ between manufacturers too!! A meter will help here.
wiring (make sure youre using the correct colums, etc)
contacts (really stick 'em in!


The resistor values should work somewhat for a run-of-the-mill NPN Si, and the suggestion of a trimmer or a pot to muck about with bias was still a good one.

This is not a complicated circuit, but sometimes your staring at it sooo hard, youre missing the very, very obvious. Hey, I once forgot to stick IC's in sockets, even to hook up the battery's - lead and took a LONG tme (2 days...) figuring that one out. Keep at it.

[R.G.]

OK, time. You two are chasing info that can best be gathered in a disciplined manner.

Can you get the info together as requested in the sticky "What to do when it doesn't work"?

[petemoore]

Here I go
  Other than what's in Debugging..which is good stuff...
I measure resistances on the board.
 With the Q1 out, I'll measure the 33k from Q1C socket lug to PS 9V.
 and everything else.
 Sometimes with this method [measureing R's in cct.] I'll get a lower reading, and figure out how that's possible through an alternate path for current to flow through the circuit besides that resistors connection points, and doing some math of such sometimes show's me where there's a wrong value resistor or bad connection.
 I use 'thumbuzz'...connecting my thumb to points in the circuit with a conductive shaft...with output plugged in and amp on, circuit powered, I start at the output jack tip, touch that...hear the buzz...then work my way through to the input...should hear a louder buzz at Q2 B than Q2C, louder yet at Q1 B.
 Test voltages and compare to those referenced at  Runoff Groove.

[Outlaws]

I did do the stuff from the sticky in a past thread, but I guess I really should do it again for everyone.


Both transistors are now 2N2369.
The 8k2 has been replaced with a 4k7 for bias purposes.
And the standard 330 has replaced the 470.


With the FUZZ pot fully clockwise
Q1:
C = 1.33
B = 0.64
E = 0.00

Q2:
C = 4.25
B = 1.31
E = 0.90

With the FUZZ pot full counter-clockwise
Q1:
C = 1.59
B = 0.66
E = 0.00

Q2:
C = 4.05
B = 1.60
E = 0.92


Not that far off of some of the stuff I saw at Runoff Groove.  There was some good info there.  I will be looking into making one of them at some point.  I like the idea for the DPDT output caps and the 1k input R to kill RF interference.

This one if at all possible though, I would like to be just a "normal" FuzzFace...if I can.   :lol:

[Mike Burgundy]

RG - that made me giggle, but you're so right. "'s sir..."
Case in point for that sticky!

Voltages don't look too far off, Q1's a bit high (what's the battery voltage?). I'm uncomfortable with the fuzz pot changing voltages - this means (to me) that either the cap attached to it is leaking or there's some kind of oscillation. Maybe it's not leaking too much - the voltage changes are in tens of volts, so it might well be nothing.
Does the  thing work now that you don't have 9V on a collector anymore? Looks like it should... Whats the problem if it doesn't?

[cjlectronics]

If you have the time, this web page has some excellent reading material on transistors and biasing (to name a few things) http://www.101science.com/transistor.htm

CJ

[Outlaws]

Okay, I just finished breadboard station (well most of it) and have all the pots and jacks set into angle iron and the breadboard in the middle.

This is making things a lot easier I tell you...no more pinched together pot connections that get loose when rotating the knob.  lol


So here is the update.


I am still getting lots of noise with the Fuzz turned all the way up.  Including some Radio waves.

I looked at all the different types of FFs out there (sili 1 and 2, tantalus, easy, NPN...etc...).

I tried adding a 10k resistor in series with my input like the Siliface 1 has.  That didn't do anything. (I realize that it also has a 100pf cap between Q1 and Q2...but I didn't try that yet)

When I took out the 10k resistor on my input, I noticed that the noise stops while playing, then comes back a second after stopping.

When rolling down on the Fuzz pot (1K) I seem to loose the noise, but I also loose all my tone with it.

So I got to thinking and went back and searched all kinds of crap and even took a couple days off.

Now look at what I notice.

Ex A:  the schematic and parts I am using.
http://www.generalguitargadgets.com/diagrams/fuzzface_lo_nn.gif

Now look at this one from AMZ

Ex B:
http://www.muzique.com/schem/fuzzface3.gif

Anything out of place?

I'll help ya...

C3

On Example A it is .01uf and on the AMZ website it is .1uf

Now I am not the smartest man in the world, but a 10th and a 100th seem like it could be a big difference.

But alas, before I go drive 10 miles to the store that isn't Radio Shack, is this even a possible fix?