The Sunface may become my white whale...

[makefast]

I'm using this layout:  http://tagboardeffects.blogspot.com/2013/03/analogman-sunface-bc108.html

I soldered the vero, and did a quick connect up to test - all kinds of fuzzy goodness.

Photos here:  http://imgur.com/a/9WT15  One day I will figure out how you guys have those super clean pedal shots, but today is not that day.

Values from the 1st Transistor (bottom of the layout, the one that the trimmer effects):
C  1.2
B  .59
E  0.0

2nd Transistor:
E  .59
C  4.4
B  1.2

So after wiring it up, initially I got no signal in bypass or active, but the LED was working.  I had the input & output wired incorrectly, fixed it at the jack and now I have bypass but the LED is not working.

When engaged, the pedal has a high pitched whine that changes in pitch with the fuzz knob.  Volume controls volume of the whine.  From Noon => 4 o'clock on the fuzz knob there is signal coming through (non-fuzzy signal), below noon nothing is audible.  Bias knob doesn't seem to effect these things.

I ran through all of the off board wiring checking continuity to/from the board and it all seems to be okay.

I may have to make an audio probe at this point, but I'm not real comfortable running an audio probe into my amp (who knows what kind of noise there may be...).  Just weird that it was working when I was using alligator clips and a testing box for offboard, but not working now.

[vitt]

To me sounds lik a faulty transistor (or oriented wrong) or could be the wiring of the fuzz pot or of the volume's one. I mean. I could have connected the output to the fuzz 2 pin. Because yours voltages are correct.

[induction]

If it worked before you boxed it but not after, then the most likely problems are bad offboard wiring, a broken solder connection, or something shorting against the pots lugs or case.

[rocket88]

Just after a quick look at your transistor voltages and Q1 is way....way......way.....did I say way off. I would throw another one in, checking orientation, and see if that fixes it. You can also try 2 other transistors in there to see if it works, then you know it's the transistor.

[induction]

rocket88 wrote

Just after a quick look at your transistor voltages and Q1 is way....way......way.....did I say way off.

Can't say I agree. Those voltages look spot on to me.

[rocket88]

Really? I thought the collector on q1 should be around 4.5v....

[induction]

rocket88 wrote

Really? I thought the collector on q1 should be around 4.5v....

That's true for Q2 and for many other BJT circuits, but not for Q1. Note that Q1's collector is tied to Q2's base, so their voltages have to be the same.

A decent rule of thumb for Q1 in a fuzz face is roughly 1 diode drop from emitter to base, and 1 diode drop from base to collector. BC108's are silicon, so 0.6 V per diode drop is pretty reasonable.

The fuzz face is an oddball among pedal circuits. Read more here.

[rocket88]

Weird. Never knew that as every one I've built fired right up so I didn't really look at it that close. Now I've got to go and check mine. Can't believe I didn't know that already, super good info to know man, thanks.

[induction]

Usually you only have to worry about the bias on Q2. Q1 is biased by voltage feedback from Q2. Notice that the base of Q1 is connected to the emitter of Q2 via a 100k resistor. The only path from Q1 base to ground is through the transistor, so Q1 base will usually be no more than 0.1-0.2 V below Q2 emitter. That means that as long as Q2 is biased correctly and Q1 is in a reasonable hfe range, we don't have to manually set the bias on Q1. So we generally focus on biasing Q2, and let Q1 take care of itself.

[makefast]

Okay,  since the Fuzz knob seemed to be controlling the whine, I pulled the pot off the pedal.  This did away with the whine, but didn't make the pedal work.

From noon to 4 o'clock the Fuzz knob passes signal, volume correctly controls volume, but the signal is faint even at the highest levels and there is no fuzz.  The pedal is like a 90% mute at this point.

I'm going to reflow solder on the pot connections and input/output as those were the only things that I changed (trimming the wires to the pots) from test box to boxing it up.

Still have no idea why the LED isn't working now unless I burned it out somehow (seems unlikely).  I'm using the PCB boards from Felix for sale in the marketplace - and it was working until I realized that the input and output were backwards (I had out going to the in jack and in going to the out jack).

[makefast]

Reflowed solder, checked with other transistors, plugged transistors back in and...it works.

It works, but is really noisy - that higher pitched whine exists from about 11 o'clock onward on the volume knob.  Not sure what this is related to.  Any thoughts/recommendations on eliminating that?  Or even how to test for what is causing it?

[makefast]

Next update - ran a big capacitor (100uf I believe) between ground and 9v rail on the board (also blew one because I wasn't paying attention to +/- and scared the crap out of my wife).  That changed the nature of the noise from a constant high pitch to sounding more like radio interference (actually like a cell phone ringing near a speaker).  The volume knob changes the intensity and the fuzz knob changes the timbre of the noise.

Googling it sounds like shielded wire for off-board hookups and/or shielding the box is the solution for this (one post noted "This is why it's called a transistor radio").  Any other solutions I haven't found?  Since I don't have shielded wire or shielding paint/foil handy, it may just be noisy until then.

[vitt]

A good way to get rid of the hiss is a 200pf capacitor between base and collector of q1 and a smaller capacitor (33pf-47pf) from base and collector q2.

[makefast]

Thanks Vitt, I'll try that next.