Help With Fuzz Face Build

[nation]

Hi All

Been racking my head over a fuzz face build over the past few days and any help will be much appreciated.

About 2 years ago I built a dual NPN fuzz face in a 1590bb. This was 2 circuits in one box, the left circuit a Ge NPN using OC140 (CV7112) and the right circuit Si using BC108. I used a blue footswitch to turn the pedal on and off which in turn connected to a DPT toggle to switch between either circuit. I used the following schematic from Dragonfly:



In addition to the stock schematic I included a switch for the 0.01uf C2 cap and a bias pot to replace the 8k2 resistor. I also increased the R1 330 ohm to circa 1k on the Ge and the Si. This build turned out great, the switching worked perfectly, it sounded great and was relatively quiet.

However, in an attempt to fine tune this build, I rebuilt this circuit last week and am now stumped by a circuit that sounds disastrous. Here are the problems I now have:

-   barely at unity gain with volume and fuzz knob on max,
-   With guitar volume on full I get a very dark unpleasant sounding fuzz,
-   With guitar volume rolled off slightly the fuzz cleans up but sounds very thin and ear piercingly bright;
-   Notes sustain but trail off abruptly into a thin unpleasant, fizzy, buzzy end,
-   The pedal sounds very high gain e.g. accidently hit the strings, a fuzzy reverberation goes through the pedal akin to hitting a reverb tank
-   Changing the 330 ohm doesn't change overall volume much.

These problems are the same for both the Ge and the Si side.

How I rebuilt the circuit?

Pulled the vero and repopulated the new board with my existing components. I kept all the off board wiring the same as I had previously but I did unwire and rewire at certain parts. I'm these went back in correctly.

I used a 2k resistor and a 10k pot in place of the 8k2 resistors. All the caps are socketed as are the transistors. I double checked all components before installing to ensure the right values were in place. Also, for the Ge circuit I had problems getting the right Q2C voltage so I replaced the 33k resistor with a 1k resistor in series with a 100k trimpot.

My voltages are as follows:

Ge:
Q1C - 0.9v
Q1B - 0.15v
Q1E - 0v

Q2C – 4.5v
Q2B – 0.9v
Q2E – 0.8v

Si:
Q1C – 1.3v
Q1B – 0.6v
Q1E - 0v

Q2C – 4.5v
Q2B – 1.6v
Q2E – 0.8v

I purchased the Ge transistors from Smallbear so assume they are in the correct range. I'm treating the red dot as the collector and using a GM328A component tester which shows the following values for the 2 Ge transistors:

Ge 1:
hFE – 75
Ic – 1.7mA
Ube – 166mV
ICE0 – 0.90mA
ICEs – 21uA

Ge 2:
hFE – 108
Ic – 1.1mA
Vbe – 130mV
ICE0 – 21uA
ICEs – 7uA

Gain for my Silicon transistors are 130and 170.

Is there anything amiss with my Ge numbers? And can anyone provide assistance on where I can begin looking to solve this problem? As the problem seems to be same for both the Ge and the Si circuit I assumed there must a problem with my offboard wiring but I have double checked this and it seems to be ok.

[thermionix]

As the problem seems to be same for both the Ge and the Si circuit I assumed there must a problem with my offboard wiring

I'm thinking footswitch or toggle switch.  Did you reuse the same ones before/after the rebuild?  Those things are easily damaged by soldering heat.  Your voltages look good to me.

Got pics?

[nation]

I'll need to get some pictures.

I did reuse the toggle and foot switch but initially I didn't unsolder any wires. I left all the off board wiring in place and just removed the wires at the point of the vero board. As the last few days progressed I moved some of the switch wires but was careful not to overheat things. I have since checked the foot switch with a DMM and the points are making appropriate contact. I will need to check the toggle switch.

[nation]

Just another note on the foot switch.

I have been using switch wiring as per the attach which came from the old Beavis website:



Does this type of wiring have any problems?

It doesn't look like the switch wiring typically seen. While this wiring has generally worked for me with other builds I have had instances of problems when builds get messier more complex.

[thermionix]

That diagram needs "ring" and "sleeve" of the input jack swapped.  Would leave battery always on as is.  Going by your posted voltages I don't think that's your problem.

[nation]

Thanks thermonix.

I just spent some more time with the circuit and have now got it fixed. I'm really not sure what the problem was but I ended up rewiring the switches and now have both sides working great. Overall volume is a little lower than expected but acceptable. unity is now at approximately 2 o'clock on the volume pot when the fuzz is on full. This is with a 470 ohm resistor replacing the 330 ohm. Other than that the fuzz sounds nice, notes trail off as expected and the guitar volume roll off cleans up nicely.

On another note, are the leakage values for my Ge transistors listed in my original post in the acceptable range for a fuzz face? Also, is the ICE0 figure the relevant figure for Ge leakage? I understand the GM328A component tester may not accurate for testing leakages for a fuzz face but if I can use the tester as base line for sorting high and low leakage transistors than that would be a start.

[thermionix]

Sorry, I'm too dumb to understand all the data from your tester.  I_CEO...current, collector to emitter...operational?  off?  on?  open?

[tonyharker]

Iceo - leakage current between Collector and emitter with base open ie not connected.

[thermionix]

I see.  900µA seems like maybe too much.

[nation]

Thanks, I saw the 900 uA I have for Q1 is high but wasn't sure whether ICE0 was the relevant figure, even moreso given I bought these transistors from Small Bear a few years back as a Fuzz Face set and assume that the leakages would have been in line at that time. I guess there may have been some drift in leakage over the years or my GM328A is just simply unreliable for leakage numbers.

I have some other Ge NPNs that I experiment with over the weekend.

[thermionix]

I honestly don't know if that Iceo is the leakage we're supposed to be concerned about.  Hopefully one of the smarter folks will chime in on this one.  But in the end, what matters is how it sounds, and you say it sounds nice, good decay, good cleanup.