Yet Another Fuzz Face Bias Question

[mrsuspend]

I recently built a germanium pnp fuzz face using the MAX1044 voltage converter (well actually I used a ICL7660S) and AC128 and A122s.
It seems to work well although I can't get it to bias lower than about -6.5 volts using a 10k trim pot. I've tried quite a few AC128s with HFEs of around 70-90 for Q1 but presumably they are a bit too leaky.
I popped an OC44 in there just to try and it biased very nicely and sounded quite good actually although I'd really like to hear a properly biased AC128 in there.
So my question is: should I go with a 20k trim pot for the bias or should I keep hunting for that perfect AC128? Is there something to be gained (ha ha, sorry) by having a spot on Q1?

Thanks!

/Magnus

[LucifersTrip]

my experience is that the closer you get to the stock 8.2K when you hit 4.5V, the better the sound is. I shoot for +/-3K (5K-11K), but, in the end, it's just up to your ear...whatever you like best

[mrsuspend]

Thanks! I guess I'll take my chances with a matched pair then 

/Magnus

[mrsuspend]

Well, I bought a set of matched AC128s AND upped the trimpot to 20k. And the thing still will not bias below around -6 volts!  
There is a bit of variation, the best one settles at around -5.6 and the worst one at about -6.8.
With an OC44 as Q1 it biases perfectly normally and also with any of around ten 2N3906s I have lying around.
So what's up, are all my AC128 and 122s leaky?
The sellers have assured me leakage should be below 300 but I haven't tested them myself....

Any advice welcome!

/Magnus

[kaycee]

Possibly something else is wrong then? Are you using stock values for the rest of the circuit? 33k from power to Q1 collector, emitter to ground. Power line resistor can be from 220R to 10K, the bigger the louder, 470R and 1k are common values and 100K feedback resistor.

You might want to double check your values, and for solder bridges if using vero. I have previously used 20k trims to bias Q2 but usually use a 3K1 in line and 10k trim now. You shouldn't be having that much trouble dialing it in with that many trannies to choose from.

[mrsuspend]

I've build almost exactly this, unfortunately this layout has since disappeared from the net:

The only differences being an ICL7660S for the MAX1044 and now a 20k bias pot.
I have to say is seems to work very well with no hum or other disturbances (although with a very specific combination of settings I do get some AM radio!) 
And it's very mysterious that the 2N3906 and even the OC44 biases so well....
I don't come from a technical background and am not sure how q1 affects q2, is it because q1 drains voltage to ground and if it's leaky it acts like a resistor, draining too little?

Some pics:

[LucifersTrip]

So what's up, are all my AC128 and 122s leaky?
The sellers have assured me leakage should be below 300 but I haven't tested them myself....

Any advice welcome!

sorry, I have to state the obvious: definitely test the leakage yourself


http://www.geofex.com/Article_Folders/ffselect.htm

[frequencycentral]

http://www.diystompboxes.com/analogalchemy/emh/emh.html

[mrsuspend]

Wow, that's exactly what I needed, thanks! 
I'll set up a test bed and feed the calculator ASAP.

Again thanks all for the help and advice!

/Magnus

[kaycee]

I can't see for sure from your picture, but the legs of Q2 need to be twisted - you have the emitter right, but the base goes in the top hole and the collector in the middle it being BCE in the layout and the tranny legs are CBE with emitter denoted by the tab in this case.

Very nice case you've made there

[mrsuspend]

I can't see for sure from your picture, but the legs of Q2 need to be twisted - you have the emitter right, but the base goes in the top hole and the collector in the middle it being BCE in the layout and the tranny legs are CBE with emitter denoted by the tab in this case.

Very nice case you've made there

Thanks kaycee, I love huge leds even though they are a complete waste of space 
Yes, Q2 has twisted legs, you just can't see it from this angle.

/Magnus

[mrsuspend]

Well, I've tested my AC128s and AC122s now (although they all took a long time to settle before giving me stable readings) and they should be usable at least for q1 (60-80 hfe).
The 122s read around 130 and 200 hfe, maybe a little on the high side.
I think I'll just breadboard a FF and try them out like I very probably should have done from the beginning! 

Thanks all!

/Magnus