FUZZ FACE volume issues

[Matej]

Hello.

I started my first project, a Fuzz Face
It's a positive ground, with AC128 Germanium Transistors... I used the first schematic - http://www.theguitarfiles.com/guitarfile503.html and added a "dying battery simulator" with a voltage meter

My problem: the sound is good, distorted as I would expect, but there is a drastic loss of volume, meaning that when it's on (with everything turned to maximum) it is noticably quieter than while bypassed..

I measured the transistors, here are the numbers 

Q1
C = 8,893
B = 9,563
E = 9,673

Q2
C= 6,651
B= 8,891
E= 9,014

It's my first thing, and i have no idea how should these numbers look like, but my first thought was that maybe the transistors are bad? The wiring should be OK, i've double-checked that.

Thanks! 

[John Lyons]

They all are high, some very high.
Q1 base is connected straight to ground so you should have 0v there
You may need to check the board and make sure the grounds are connected
at all the ground points, and connected to the sleeve of the input jack and battery
red wire.

john

[Renegadrian]

I built a BC108 version (the smaller silicon vero layout) and it calls for a 100k volume pot while others have a 500k - I don't know if changing the pot would change the output, but it has a slight volume loss...Not so dramatic, but you can tell there is...

[yeeshkul]

Also when you sort out the issue with voltages, go for a bigger output divider resistor - 470->1-1.5k works fine and adds reasonable volume. The divider is what causes the volume drop - but not in this case ... judging by the voltages.

[Matej]

okay, i checked it again....i measured it wrongly first time....god im stupid 

anyway

Q1
C = 0,7961
B = 0,1188
E = 0

Q2
C= 3,186
B= 0,7521
E= 0,6312

hope that's better

[Curly]

looks like your bias is a bit low ... you'd like the Q2 collector to be around 4.5-5V

however, it would be nice to know the gains of the transistors

this is one reason why trimpots are handy for biasing ... so you can really dial it in

[yeeshkul]

Voltages are fine. Change the 470 res to 1k5 and adjust the bias resistor so there is about 5V on the Q2 collector. Fuzz Face doesn't use clipping diodes for fuzz tone so you have to balance output volume and fuzziness until you're happy.

EDIT: the higher bias voltage on Q2 you have (let's say up to 7V), the more volume you get but no juicy fuzz. If you use a simple dying battery sim. (=just a bare pot) the resistance will add itself to the both collector resistors and volume will go up by paradox.

Fuzz Face by it's nature is a quiet, fuzzy pedal.

[Matej]

okay, i biased the collector on 4,7V or so, and changed the 470R resistor, but still, volume drop is significant plus now it sounds very "destroyed", like it had been bit-crushed (if you can imagine what i'm talking about  )

i could get along with the sound as it would be interesting to have something different (maybe) in the chain, but now it's unusable because of the volume drop (i mean it's still A LOT quieter than when bypassed)

[yeeshkul]

Use an audio probe to find out where the volume loss is happening. FF is a quiet abut not dead quiet pedal.

[petemoore]

  Pick a node, note orientations of polarized components, count total number of components at that node.
  Measure every resistor...
  Pick another node, repeat the 'dogmatic' approach on this one also.
  Do that for all the nodes.
  Clip to the node and test that everything that is supposed to be connected, is, and that there aren't any stray connections [a bit more tricky...finding the 'secret, hiding' miswire.
  Audio probe is a good one too, this should tell you a capacitor is passing AC [and you can test for DC blocking capability with the DMM], though bad caps [even used ones] are very very rare, electrolytics being more prone to damage from age or reverse polarization, still very rare, just resigning to "it's probably not a bad capacitor" seems to work well, unless you're using really old electrolytics etc.
  Anyway, there's something in there that shouldn't be or something else, the FF should have a big boost over unity.
  When I was building FF's, I would routinely place a 470k...where the 470 or 1k would be, measureing every resistor value and that it actually connects to 'x' and 'y'...should help to set up the bias, then it's just the signal path flow...