Fuzz Face Debug Help Needed

[RedHorseMane]

Project Name: Fuzz Face
Sources
•    You can Build the Perfect Germanium Fuzz Face (electrosmash.com)
•    The Technology of the Fuzz Face (Geofex)
•    Picking transistors for FF Clones (Geofex)
•    Positive Power for the PNP Fuzzface (muzique.com)
•    LED Popping (muzique.com)
•    What really causes switch pop (mrblackpedals.cxom)
•    Reverse Current Protection Using MOSFET and Comparator to Minimize Power Dissipation (Texas Instruments)
Methodology
I followed the advice and schematics in the above literature, and other sources, as closely as I could. I fist made the circuit using Veroboard. Once I had solved some small issues, I had a PCB made. When designing the PCB, I changed the 32k and 8.2k resistors for a 50k and a 10k trimmer resistor respectively. I also added the reverse polarity protection and corrected errors that I made on the Veroboard regarding the pull-down resistors.
The PCB circuit works, in that there is sound at a  reasonable volume, but no fuzz. I compared the output of the Veroboard circuit to the PCB and found that, although the internal signals were all similar, the signal beyond the output capacitor was almost unchanged on the PCB, and quite different on the Veroboard.
In diagnosing, I checked all the voltages and found them to be close to required. I then began to check the individual components, eventually swapping out all the capacitors, which made no discernable change. I disconnected the pull-down resistors and swapped the transistors for others, which made no difference. I have checked the polarity of the capacitors, measure the resistance of all the resistors and probed another PCB from the batch to check for errors on the board, all without finding the problem.
Parts Substitutions
I have used 2N527 Ge PNP transistors, which I matched by testing their Hfe and leakage. In the Veroboard circuit, I used mainly cheap (Bojack) parts, but upgraded the parts for the PCB. During my investigation, I desoldered the new capacitors and replaced them with ones identical to those on the Veroboard version. There was no significant change.
Negative Ground
Following advice I found at 'Positive Power for the PNP Fuzzface (muzique.com)', I changed the circuit from positive to negative ground so that the pedal would be compatible with the others on the board and could use a similar power supply. That had already been done on the Veroboard version, which works, so there is no change from that to the PCB.
I have used a 470 resistor for R4 and 22u capacitor for C2. Apart from those, the values are as published on electrosmash.com, and many other sources.
What is not Working
No fuzz!
I have included a schematic, the voltages and pics of the scope readings taken at the audio out jack of the respective boards.
Any advice and input will be much appreciated.














   

[duck_arse]

welcome to the forum.

I see a problem. Q1 collector is connected to Q2 base, so those two pins must measure the same. your's don't. have you written them down wrong, or fitted the transistor wrong.

we always like to see photos of what we are debugging. oh, and R6 is redundant.

[R.G.]

Yep, Q1 collector and Q2 base should be identically the same voltage.

Another issue I can see is that the collector of Q1 is sitting at the 9.04V that is the PNP transistor's
ground". Is there a mistake in the voltage list, or have I simply not had enough coffee yet to wake up? Perhaps the "C" and "E" are reversed?

Also, the oscilloscope pictures show signals that should sound incredibly fuzzy. Is there maybe a mistake in your bypass wiring? I have not yet traced the schematic for the bypass switch.

[RedHorseMane]

Yep, Q1 collector and Q2 base should be identically the same voltage.

Another issue I can see is that the collector of Q1 is sitting at the 9.04V that is the PNP transistor's
ground". Is there a mistake in the voltage list, or have I simply not had enough coffee yet to wake up? Perhaps the "C" and "E" are reversed?

Also, the oscilloscope pictures show signals that should sound incredibly fuzzy. Is there maybe a mistake in your bypass wiring? I have not yet traced the schematic for the bypass switch.

My apologies; finger trouble. I reversed the collector and emitter voltages, as you guessed. The emitter is 9.04 because I am using a bench top power supply and don't seem to be able to set it to 9V exactly.
The signal with the sharp spikes is from the Veroboard circuit, which sounds very fuzzy through the amp. The signal from the PCB, which is much more rectangular, does not produce much distortion through the amp at all. (Volume and Fuzz turned all the way up on both.)
Apologies for sending you down the wrong path with my incorrect table.

[RedHorseMane]

welcome to the forum.

I see a problem. Q1 collector is connected to Q2 base, so those two pins must measure the same. your's don't. have you written them down wrong, or fitted the transistor wrong.

we always like to see photos of what we are debugging. oh, and R6 is redundant.

Thank you.
I transposed the collector and emitter of Q1. I thought I'd checked it carefully, but obviously missed the error, which is probably what's happened with the circuit.
Is R6 redundant because the DC signal has already been filtered out at that point? (I hadn't thought of that before.)
I'll post a picture of the PCB as suggested.
Thank very much for taking the time to help out.

[antonis]

Is R6 redundant because the DC signal has already been filtered out at that point?

R5 & R6 serve as anti-pop (pull-down) resistors for C1 & C3 respectively..

R6 is retundand due to RV1 presence..

P.S.
Welcome, also..

[mozz]

If you are having problems, go back to the regular positive ground version and get that working first.

[RedHorseMane]

Here's a photo of the PCB. It's a mess right now because I've been swapping out capacitors and removed R5 & R6. Sockets for transistors are way too close together because I forgot about the packages...

[RedHorseMane]

If you are having problems, go back to the regular positive ground version and get that working first.

That would be sensible, but unfortunately I would have to redesign the PCB. The Veroboard version is negative ground and from the same schematic, so the design is okay; it's the execution that's lacking...

[antonis]

Collateral note: Is R7 really 470k..??

[RedHorseMane]

Yes. Unsurprisingly, the LED is a little dim. I think I might change that when I get the problem sorted. :-)