News:

SMF for DIYStompboxes.com!

Main Menu

Fuzz face problem

Started by Coelhomatias, January 08, 2019, 12:39:08 PM

Previous topic - Next topic

Coelhomatias

Hi guys!
So yesterday I built a fuzz face based on this schematic
http://prntscr.com/m4jy8y
I used BC108C transistors and not the ones on the schematic, that was only to get the footprint.
However the circuit does not work as intended, the bias trimmers have to be set to minimum to get a good sound and when the fuzz pot gets below 9 o'clock it outputs no sound at all. Can you help me pls?


antonis

<off topic ON>

Delete D1 or D2

<off topic OFF>
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

j_flanders

With that little info, the usual reply, and solution, is: check if you haven't put the transistors in backwards.
If that didn't help, post the collector, base and emitter voltages of both transistors.

Coelhomatias

Quote from: antonis on January 09, 2019, 06:24:07 AM
<off topic ON>

Delete D1 or D2

<off topic OFF>

I can but i dont think that is the problem, i used this method of filtering and reverse polarity protection in other pedals and all worked very well

Coelhomatias

Quote from: j_flanders on January 09, 2019, 07:18:54 AM
With that little info, the usual reply, and solution, is: check if you haven't put the transistors in backwards.
If that didn't help, post the collector, base and emitter voltages of both transistors.
Thank you for your help, no the transistors are not reversed i checked for that, as soon as i can i will post the voltages

antonis

Quote from: Coelhomatias on January 09, 2019, 07:58:38 AM
I can but i dont think that is the problem,
Of course it isn't the problem - that's why I marked my suggestion as "off-topic"..  :icon_wink:

Placing both D1 & D2 is an overkill..
(not to say that D1 counteracts D2 on reverse polarity protection..)
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

Coelhomatias

Ok so i just measured the voltages on both transistors,

Q1:
E - 0V
B - 0.6V
C - 1.6V

Q2:
E - 0.95V
B - 1.6V
C - 4.5V

These measurements were taken with the fuzz pot halfway, but if i start to decrease the fuzz the voltages decrease as well, at minimum there are only mV in C of Q2. The effect is bigger in Q2 than Q1

pinkjimiphoton

i see a few things. the diode isn't gonna hurt anything, but it WILL cause a .7volt drop in your supply voltage which will impact the biasing.
bc108's are weird for fuzzface. i know people use them, but too high gain usually. that can be fixed by adding resistanced from e to ground on both q's

10n for an output cap is pretty small, and won't let most of the lower range of your guitar thru. can make it sound kinda stuffy.

imho 220p as a snubber is likely the problem, that is WAY too big and will wipe out much of your highs. 100p is a lot. try 20-47p there, but only if its needed. the 47p on q2 should be plenty.

r6 has absolutely no function at all, and shows it tied to the input of the circuit which is also not right. likely whats @#$%ing up your bias.

i'd go with the standard b500k output pot.

when biasing, use your ears. your voltages are in the ballpark pretty much. but do it this way... crank the fuzz pot full, then adjust your bias til it sounds ok. check by rolling your guitar back. you want it to clean up substantially. when you have it right, about half way up should sound roughly like its bypassed but with a bit more sparkle and compression.

trying to bias it half way up its gonna change as soon as ya start playing thru it.

thats why the fuzz pot cuts off at 9:00 most likely. you want a 1k a or c taper there, a b/linear taper will give ya weird results too.

there is a fuzzface resistor calculator that joe davidson wrote years ago somewhere on this forum... if ya do a search, you will find it.  you input the gain of your q's and it tells ya which resistor values will work best for what you have.
don't be surprised if ya have to change out the 470r to either a lower (220r, 330r) or higher value (560r. 680r, 1-2k) to get it to sound right.

i'm betting the betas are too high on the q's.

ya want an hFE of around 70 for q1 and 100 on q2, =/- 20% tolerance or so... the bc108's are probably way higher than that.

try adding a 1k resistor from e to ground on q1, and maybe 470r from e to the input of the fuzz pot on q2. you may need to play with the resistor values, but that way you can limit the gain to the sweet spot range ya need.  this sets the gain.. adjusting from c adjusts the bias, which is different.

between the two, ya can "float" the transistor in a nice range that has useful operation for this circuit. hope it helps man... peace



Quote from: Coelhomatias on January 08, 2019, 12:39:08 PM
Hi guys!
So yesterday I built a fuzz face based on this schematic
http://prntscr.com/m4jy8y
I used BC108C transistors and not the ones on the schematic, that was only to get the footprint.
However the circuit does not work as intended, the bias trimmers have to be set to minimum to get a good sound and when the fuzz pot gets below 9 o'clock it outputs no sound at all. Can you help me pls?
  • SUPPORTER
"When the power of love overcomes the love of power the world will know peace."
Slava Ukraini!
"try whacking the bejesus outta it and see if it works again"....
~Jack Darr

pinkjimiphoton

  • SUPPORTER
"When the power of love overcomes the love of power the world will know peace."
Slava Ukraini!
"try whacking the bejesus outta it and see if it works again"....
~Jack Darr

Electric Warrior

Quote from: Coelhomatias on January 10, 2019, 02:19:47 PM
Ok so i just measured the voltages on both transistors,

Q1:
E - 0V
B - 0.6V
C - 1.6V

Q2:
E - 0.95V
B - 1.6V
C - 4.5V

These measurements were taken with the fuzz pot halfway, but if i start to decrease the fuzz the voltages decrease as well, at minimum there are only mV in C of Q2. The effect is bigger in Q2 than Q1

There must be something wrong here. The fuzz pot shouldn't make much of a difference for the voltages. Maybe it's miswired?

Once you have eliminated this issue, you can take a closer look at biasing.
The voltages aren't far off in its current state. You set Q1C a little high. Try setting it between 1.2 and 1.25 V. After than you will need to set Q2C again as Q2 is biased by Q1. Between 4.5 and 5.5 V should get you a resistance on Q1C that should be fairly close to the stock 8k2 if Q1C is set correctly.

Quote from: pinkjimiphoton on January 10, 2019, 03:36:26 PM
i'm betting the betas are too high on the q's.

ya want an hFE of around 70 for q1 and 100 on q2, =/- 20% tolerance or so... the bc108's are probably way higher than that.


I disagree. Dallas Arbiter mostly used BC108Cs (i.e. from the highest gain bucket) in these, so high gain transistors will probably give most people the tone they are expecting from a silicon Fuzz Face.

The gain actually helps with the bias in a stock circuit. With too little gain it may need to be tweaked.

Quote from: pinkjimiphoton on January 10, 2019, 03:36:26 PM
10n for an output cap is pretty small, and won't let most of the lower range of your guitar thru. can make it sound kinda stuffy.

(...)

i'd go with the standard b500k output pot.

10n (0.01µ) is the most common stock value, which should provide plenty of low end in combination with the stock 470k pot – too much for many guitarists, actually. 250 k will tighten it up a little..

Quote from: pinkjimiphoton on January 10, 2019, 03:36:26 PM
r6 has absolutely no function at all, and shows it tied to the input of the circuit which is also not right. likely whats @#$%ing up your bias.

Actually it's shown to connect to a pin, but not which pin. :icon_question: confusing..

pinkjimiphoton

Quote from: Electric Warrior on January 13, 2019, 07:34:45 AM
Quote from: Coelhomatias on January 10, 2019, 02:19:47 PM
Ok so i just measured the voltages on both transistors,

Q1:
E - 0V
B - 0.6V
C - 1.6V

Q2:
E - 0.95V
B - 1.6V
C - 4.5V

These measurements were taken with the fuzz pot halfway, but if i start to decrease the fuzz the voltages decrease as well, at minimum there are only mV in C of Q2. The effect is bigger in Q2 than Q1

There must be something wrong here. The fuzz pot shouldn't make much of a difference for the voltages. Maybe it's miswired?

Once you have eliminated this issue, you can take a closer look at biasing.
The voltages aren't far off in its current state. You set Q1C a little high. Try setting it between 1.2 and 1.25 V. After than you will need to set Q2C again as Q2 is biased by Q1. Between 4.5 and 5.5 V should get you a resistance on Q1C that should be fairly close to the stock 8k2 if Q1C is set correctly.

Quote from: pinkjimiphoton on January 10, 2019, 03:36:26 PM
i'm betting the betas are too high on the q's.

ya want an hFE of around 70 for q1 and 100 on q2, =/- 20% tolerance or so... the bc108's are probably way higher than that.

Quote
I disagree. Dallas Arbiter mostly used BC108Cs (i.e. from the highest gain bucket) in these, so high gain transistors will probably give most people the tone they are expecting from a silicon Fuzz Face.

The gain actually helps with the bias in a stock circuit. With too little gain it may need to be tweaked.

well, i disagree. to me, the silicon ones with high gain sound like shit, and don't represent what the circuit can actually do. but if ya LIKE that sound, all the power. i don't. to me it constricted, too nasal and doesn't clean up right.



Quote from: pinkjimiphoton on January 10, 2019, 03:36:26 PM
10n for an output cap is pretty small, and won't let most of the lower range of your guitar thru. can make it sound kinda stuffy.

(...)

i'd go with the standard b500k output pot.
Quote
10n (0.01µ) is the most common stock value, which should provide plenty of low end in combination with the stock 470k pot – too much for many guitarists, actually. 250 k will tighten it up a little..


stock value is 100n, 10x the size of 10n. limiting the bass output here will strangle a good bit of saturation and thunk.**** 250k will diminish the overall output, as well as attenuate a bit of the tone some. you can go whatever value you like, but it will affect the final primary voice of the fuzz significantly. that said, depends largely on how you use it.
if you're a bedroom player with the fuzz pot dimed and the volume barely on with your emg equipped superaxe with floyd dimed as well, yeah, you may like that.
if ya make a living with the things tho, ya may find the way "most people" do things has very little bearing on actual reality, so ;)
too thin won't get ya there either. like everything else, its a compromise. when you change one thing, it will affect other things, particularly in this circuit.



Quote from: pinkjimiphoton on January 10, 2019, 03:36:26 PM
r6 has absolutely no function at all, and shows it tied to the input of the circuit which is also not right. likely whats @#$%ing up your bias.
Quote
Actually it's shown to connect to a pin, but not which pin. :icon_question: confusing..

if ya look close, it shows the input of the input cap as the other connection to the hanging resistor...  i was like... huh? if ya do that, how can it possibly work?

in this circuit, i've learned that there's a lot more going on than ya would ever expect ;)


***** haha jesus, i've built 100's of these @#$%ers, just now noticed the output cap is 10n not 100n, thanks EW!!! lmao  now i know why they're so phat ;)
  • SUPPORTER
"When the power of love overcomes the love of power the world will know peace."
Slava Ukraini!
"try whacking the bejesus outta it and see if it works again"....
~Jack Darr

Electric Warrior

Quote from: pinkjimiphoton on January 13, 2019, 01:08:57 PM

well, i disagree. to me, the silicon ones with high gain sound like shit, and don't represent what the circuit can actually do. but if ya LIKE that sound, all the power. i don't. to me it constricted, too nasal and doesn't clean up right.

(...)

250k will diminish the overall output, as well as attenuate a bit of the tone some. you can go whatever value you like, but it will affect the final primary voice of the fuzz significantly. that said, depends largely on how you use it.
if you're a bedroom player with the fuzz pot dimed and the volume barely on with your emg equipped superaxe with floyd dimed as well, yeah, you may like that.
if ya make a living with the things tho, ya may find the way "most people" do things has very little bearing on actual reality, so ;)
too thin won't get ya there either. like everything else, its a compromise. when you change one thing, it will affect other things, particularly in this circuit.

(...)

if ya look close, it shows the input of the input cap as the other connection to the hanging resistor...  i was like... huh? if ya do that, how can it possibly work?

(...)

in this circuit, i've learned that there's a lot more going on than ya would ever expect ;)

(...)

***** haha jesus, i've built 100's of these @#$%ers, just now noticed the output cap is 10n not 100n, thanks EW!!! lmao  now i know why they're so phat ;)

Silicon Fuzz Faces sure are a different flavour, but I guess they have their place. If you want them to sound like a germanium one, the circuit definitely needs tweaking. I simply use germanium transistors for THAT tone.  ;D

Input, output, ground, 9V and that resistor all have connections to "pin", but they're not all meant to connect with each other. What is it that tells you that the resistor connects with the input? Am I missing a detail here? ???

Some schematics for silicon fuzz faces do have a 100n cap at the output, but I think that's a typo rather than a rare variant. Or maybe someone misinterpreted the color bands of a tropical fish cap..  ;D

With the 10n the difference the Level pot makes should be much more dramatic. Even a 100k there can work well, but certainly is less "Fuzz Face" flavoured. My MK1.5 Tone Bender has that value and it sounds great. The low end is tighter than with a larger pot, but not so much that it gets thin sounding. Just less boomy. The big low end of a Fuzz Face can make things muddy in a mix with bass guitar and drums. But yes, it interacts a little differently with the amp as it doesn't push it as hard at lower frequencies so you get less saturation. Depending on your amp, its headroom, your settings and the volume you're playing at, this may or may not be a good thing. It's not necessarily a bedroom/stage setup issue.

It's funny, but it's still possible to obeserve things about this circuit that haven't been discussed before..

pinkjimiphoton

bro, try a highpass filter before the input. fuzzfaces make KILLER overdrives that way ;)

yeah. every single time i make one of these, it ends up being slightly different.
i'm still chasing ge tone on si devices...

it can happen... fleetingly... until ya change just ONE thing somehow ;)

my latest trick is messing with the gain of each q to get 'em in the ballpark and then biasing accordingly. i think if ya get a si to where its beginning to saturate a good bit without hard clipping you can get pretty close to a ge tone... jumping thru various other hoops, of course.

hey man.. happy new year, btw.. peace out
  • SUPPORTER
"When the power of love overcomes the love of power the world will know peace."
Slava Ukraini!
"try whacking the bejesus outta it and see if it works again"....
~Jack Darr

Coelhomatias

Hey guys actually r6 is for the LED it has nothing to do with the circuit, it's not connected to the input, its to the 9V rail. I read your input and i'm going to test some things. I did a pcb out of it so modifications are not that easy. The schematic and values were used from fuzz dog's kit. Thank you for the great inpt guys!!

pinkjimiphoton

the input, output, bottom of that resistor and even ground have circles that say
"pin" in them...  weird!
good luck tho, you'll get it. once ya get it running, you'll be psyched ;)
  • SUPPORTER
"When the power of love overcomes the love of power the world will know peace."
Slava Ukraini!
"try whacking the bejesus outta it and see if it works again"....
~Jack Darr

Electric Warrior

Quote from: Coelhomatias on January 14, 2019, 02:40:07 PM
I read your input and i'm going to test some things.

you should add the orientation of the 22µF cap to your list of things to test. If the wiring of the fuzz pot turns out to be correct, this might be the culprit..