My Fuzzface requires a signal stronger than my guitar...

[pinkjimiphoton]

@electric warrior... so in this case, the transistors are pnps being run reverse beta?

or am i still missing something? i see it's still a + ground circuit for all intents... you're right, i'm not getting it.

every fuzz i've ever built, pnp or npn, the c is always higher than the e. (not necessarily "positive")... looking at that schematic i can see where it's backwards and the voltages should be backwards as well...

i think. (barely, apparently  )

looks to me in my limited experience like it would make it motorboat like crazy, and probably shut the transistors off for all intents.

i gotta google up more info... thanks man

[dwmorrin]

If it's connected correctly, turning the pot should have no effect on the dc voltages.

[Electric Warrior]

@electric warrior... so in this case, the transistors are pnps being run reverse beta?

or am i still missing something? i see it's still a + ground circuit for all intents... you're right, i'm not getting it.

every fuzz i've ever built, pnp or npn, the c is always higher than the e. (not necessarily "positive")... looking at that schematic i can see where it's backwards and the voltages should be backwards as well...

i think. (barely, apparently   )

looks to me in my limited experience like it would make it motorboat like crazy, and probably shut the transistors off for all intents.

i gotta google up more info... thanks man

Actually ground is negative. That's the point of the mod.
I never tried it, but motorboating, oscillation and all kinds of hum/noises do seem to be common issues when wiring a PNP Fuzz Face up like this.

[joakim]

Well, this is all becoming very complex for me...
Anyway, hooking up the 8.8V battery, the voltage at Q2's emitter is 8.6V.

Is the common conception that the project should be reworked into a "normal" Fuzzface? I've forgotten why I opted for the "positive power supply" version in the first place...

[dwmorrin]

The emitter voltage by itself is unimportant.
You need to compare the base voltage to the emitter voltage.
When the transistor is on, you will get your 140mV drop across the transistor base and emitter junction.
Check the 33k, and the 100k.
You could also try the circuit without Q1 to see if q2 behaves normally then.

[PRR]

http://bildr.no/view/amp6SXJ0

P1 100K is wrong. Should be 1K.

100K is starving the heck out of it.

[Electric Warrior]

Is the common conception that the project should be reworked into a "normal" Fuzzface? I've forgotten why I opted for the "positive power supply" version in the first place...

So it can be daisy chained with other negative ground pedals?

[joakim]

http://bildr.no/view/amp6SXJ0

P1 100K is wrong. Should be 1K.

100K is starving the heck out of it.

Sorry bout this, but there is an error in the notation in the schematic: P1 is indeed 1k (as mentioned in my previous post). I would like to update the graphic in the original post so as to prevent further confusion, not sure if that's possible.

Thanks tho :-D you're spot on to notice it!

corrected schematic:

[dwmorrin]

What is R3's value on the circuit board?  Kinda looks like 510Ω.

If you've checked everything, and all is well (at least for resistances, connections), then perhaps it is time to try increasing R2.
We're running out of things to check.

Edit: Perhaps easier is to clip in some resistance across the 33k (lowering the resistance) to turn on Q2. Less invasive test.

[joakim]

I'm sorry if I'm being a pain, my knowledge on electronics is very limited... which is why I'm so devoted to finishing this project   It's a learning process!
So, I've measured the voltages on Q2 again, and, when VCC = 8.76V, I get (relative to ground):

V_E = 8.67V
V_B = 8.64V
V_C = 0.66V

I desoldered one leg of the resistors, and checked them with the dmm:

R1 = 32.8k
R2 = 8.16k
R3 = 509
R4 = 99.0k

Regarding C1: The original schematic has this as an eletrolytic cap. However, I couldn't find an electrolytic 2.2uF cap at the time, and I was also told that this would make no difference anyway... which is why C1 is not polarized.

On a side-note, I noticed another error in my old Eagle file: the labels "Q1" and "Q2" are swapped, wrt the labelling in the photo of my PCB. Since Q1 and Q2 are identical components, this doesn't really matter much... until I started discussing the PCB and circuit in this forum! Anyway, every time I've refered to Q1 and Q2 in this thread, I've kept consitent with the photo. So, here's yet another updated schematic:
 

[PRR]

> when VCC = 8.76V, I get (relative to ground):
> VE = 8.67V
> VB = 8.64V
> VC = 0.66V

Not your fault: upside-down polarity makes odd-looking values.

In *this* case, voltages relative to Vcc are more-like what we are used to seeing with conventional polarity.

VCC = 8.76V (relative to vcc)
VE = 8.67V == (0.09V)
VB = 8.64V == (0.12V)
VC = 0.66V == (8.09V)

We expect voltage from B to E to be (for Ge) 0.1V to 0.3V. You have 0.03V. This seems far too small. It suggests that Q2 is not getting turned-on by Q1.

Can you put the Red meter lead on "+9V", then probe *seven* points? The three leads on each transistor and the Ground. Post that.

[joakim]

> when VCC = 8.76V, I get (relative to ground):
> VE = 8.67V
> VB = 8.64V
> VC = 0.66V

Not your fault: upside-down polarity makes odd-looking values.

In *this* case, voltages relative to Vcc are more-like what we are used to seeing with conventional polarity.

VCC = 8.76V (relative to vcc)
VE = 8.67V == (0.09V)
VB = 8.64V == (0.12V)
VC = 0.66V == (8.09V)

We expect voltage from B to E to be (for Ge) 0.1V to 0.3V. You have 0.03V. This seems far too small. It suggests that Q2 is not getting turned-on by Q1.

Can you put the Red meter lead on "+9V", then probe *seven* points? The three leads on each transistor and the Ground. Post that.

Ok, I think I get your drift -- this is to do with the "positive power supply" inversion mod, right?
To the best of my measuring knowledge, I get the following readouts:

Q₁
V_C = 71.6mV
V_B = 53.9mV
V_E = 0  (since this is also VCC)   

Q₂
V_C = 71.3mV
V_B = 71.3mV
V_E = 62.0mV

GND = 8.76V

[dwmorrin]

Going back to page 1 of this thread...
Yes, we know that Q1 is saturated, and Q2 is cutoff.
Here are two simple in-circuit tests you can do with alligator clips:
1) Clip Q1 base to Q1 emitter.  Now record Q2 voltages.  Does the collector voltage move?  Does the BE voltage look like a Ge diode drop now?
(Unclip Q1 base to emitter for #2)
2) Clip a 100k resistor across the 33k resistor leads (put a 100k resistor in parallel with the 33k).  Does the Q2 collector voltage move?

Edit: Wow, your last readings of Q2 are totally different.  The base and collector are the same voltage now. Can you confirm this?

[joakim]

Edit: Wow, your last readings of Q2 are totally different.  The base and collector are the same voltage now. Can you confirm this?

Going nuts:
Hooking up the dmm in the exactly same way, I now get the following readings:   
(VCC is 8.74V)

Q₁
V_C = 84.3mV 
V_B = 58.5mV
V_E = 0mV

Q₂
V_C = 8.09V  <--- NB!
V_B = 82.6mV 
V_E = 65.8mV   

... waiting a few minutes, and reading again, I get

Q₁
V_C = 77.5mV
V_B = 56.2mV
V_E = 0mV

Q₂
V_C = 8.07V
V_B = 77.1mV
V_E = 63.4mV

[dwmorrin]

Those voltages are consistent, and we're back on track.
Don't mind the little changes.  In this circuit, ambient temp and your warm fingers can change the voltages.

Do you get what I'm saying about trying to turn Q2 on?
We've checked your transistors.
We've checked resistors. We've checked connections.
It's OK to change resistor values if you haven't made some major error.
Lowering the 33k will allow Q1 to move a bit away from saturation, and allow Q2 to turn on.
You can check for this by adding a resistor in parallel.  If you have clips, you don't even have to desolder anything, which is why I would recommend it before messing with other resistors.

[PRR]

All this 80mV stuff is too low.

This *could* be because the Q1 Collector resistor (typically 33K) isn't really connected. Bad solder joint etc. Bad resistor can happen but is very unlikely. I like to do things like confuse 33Meg for 33K.

[joakim]

I was thinking to add an ~16k resistor in parallel with R1, to cut the resistance to a third. Unfortunately, I don't have clips, but my PCB is already quite a disaster area after desoldering transistors, resistors
About the transistors (turning Q2 on etc), I'd love it if you'd give a brief explanation on what the voltages indicate, and why that means Q2 is not turned on... (hopeful)

Thanks again, and I'll report back later after done soldering always takes more time than expected.

[dwmorrin]

For BJT transistors, the base-emitter junction and the base-collector junctions are diodes.  You may need to read about diodes and their forward voltage drop, if you're not familiar.
When you apply voltage to the base, it should "come out" the emitter minus one diode drop.
Ge diodes drop 100-300mV (or 0.1-0.3V) usually.  Si is usually around 600mV (or 0.6V).

The quickest test you can make of a BJT transistor in circuit (in my experience) is to measure the voltage from collector to emitter, and then the base to emitter voltage.  That is not measuring from ground.  That is red lead to collector, black to emitter, and then the second test is red lead to base, black lead to emitter.

The first test tells you if the transistor is cutoff (high voltage), saturated (low voltage), or active (somewhere in between).
The second test tells you why this condition is present.  Less than a diode drop means cutoff.

When Paul says "all this 80mV stuff is too low" he means we expect to see a good diode drop (>100mV) across the base-emitter junction.

I had a feeling your board was getting chewed up.  Clips are real handy. I would suggest getting some.
See if you can manage to just poke a wire to get Q1's base to ground.  If Q2's voltage moves, then Q1's saturated state is to blame.

I also checked my parts drawer and found a Ge pnp fuzz face kit from smallbear that suggests very different resistor values to get it to work.   I think when you use oddball Ge transistors, you can't always stick with the stock values.

[joakim]

I tried taking a wire from Q1 base to GND, but Q1 just got real hot...  
So I only measured Q2 collector (relative to VCC, as above) to 0,75V. The base of Q2 started out at around 100mV, and started rising (until I realized something was smelling). Hope I didn't kill it  
Maybe I shorted something out... Anyway, without clips, I was not successful in connecting Q1b to Q1e, nor Q1b to GND. I could try again tomorrow, but then I'll need to find some better solution.

However, I did put a socket for R1, so I could replace it with a 15k resistor, which gave the following readings:

VCC = 8.84V

Q₁
V_C = 7.7V
V_B = 82.6mV
V_E = 0

Q₂
V_C = 3.1V
V_B = 723mV
V_E = 636mV

[Seljer]

"Ground" in your case the the +9V because you've got a slightly unorthodox layout. If you connected the base it to 0V, you let all the current that could run through the base-emitter junction, basically a short circuit which is why it got hot  

As you've now got things socketed, you can now easily swap the transistors around and ti see if just one is acting funny. You can also try putting in a silicon pnp transistors to see if there is something wrong with your layout.