Pastyface Biasing Issues

[tenken]

Hey guys,

I'm having a bit of trouble on a pasty face build, trying to do a Soulbender with a set of transistors from Smallbear (Q1 - NQ6828, Q2 - OC76, Q3 - OC76).  I've been mucking around with it for days, trying some AC125 and AC128s I have, even rebuilt it on breadboard to solve the problem.

The issue I've got is biasing Q3.  If I set it up the way I think it is supposed to be, the voltage on the collector is typically between -0.11V up to -1.5V depending on the leakage of the transistor and trim pot.  The lower the leakage, the higher I can get the voltage on the collector (more negative).  But yeah, on the recommended transistor from Small Bear (OC76, G = 101, L = 267) the range over the sweep of the trim pot is -0.11 to -0.5V.

The way I've got it set up for the OC76 is red dot = collector, which connects to R8, R7 and C6.  In my frustration, I flipped the transistor, i.e. ground the red dot and all of a sudden I can bias from 0V to -7.6V and seems to make sense.  I tried this with an AC125 as well and got the same story (biases over 0 to -7.6V by connecting the side closest to the tab to R8, R7 and C6 - which I feel is wrong).

I played it with the way I thought it was supposed to be (red dot = collector = not grounded) and it sounded good, but didn't clean up particularly well with lower pick attack/guitar volume.  I'm a bit worried to play it the other way around in case I break something.

Q1 and Q2 have -3.5V and -1.5V on their collector (red dot) and emitter respectively, inserted how I would expect.  I also tried a few different germanium diodes, and got a similar story when I reverted the design to that which came with the transistor set from Small Bear).

I was wondering, what do you guys think?  What range does your pasty face bias over (I've read you are supposed to bias Q3 to -4.5V or maybe even -7V)?  Which way would you suggest I insert the transistor?  Am I better off leaving it in the way I think is right with a bias of -0.5V or switch it around and bias to something more normal?  Am I just an idiot, and have completely misunderstood?

Sorry for the long post, but I also wanted to thank everyone.  I've got a bit of an addiction to madbean pedals (this is my 25th build) and reading past posts has helped me solve a whole bunch of problems in the past.

[Bret608]

Hi,

I had a similar issue when I built my Pastyface this summer. I ended up doing a lot of reading over at DIY Stompboxes. Long story short, don't expect this to bias to 4.5v like a Fuzz Face as that's not what the circuits needs. Historical versions of this did not use the trimmer/10k resistor setup that the Soulbender uses. They had leaky transistors for Q3. I asked someone with two vintage units to measure the Q3 collectors--they got 2 volts on one and 2.8 on the other.

With the trimmer in the Soulbender version, you'll probably never get over 1.5 volts, but you don't need to. Mine sounds good anywhere from 1.3 volts (with 500k in there a la the trimmer) to 3.6 volts (for which I had no trimmer and its pads jumpered, and the 10k resistor next to the ge diode empty). It ended up sounding best at 1.9 volts, which I did by leaving off the trimmer and using 1 meg for the resistor (also, I used the Sola Sound part values, but the logic with the biasing is the same).

My Q3 transistor wasn't very leaky, and that's the crux of the issue. MkIII Tonebenders are known to sound better with leakier transistors for Q3, so the point of the Soulbender trimmer is that it allows you to simulate some leakage. I think the circuit starts to sound good at about 1.3 volts on Q3's collector. If it's as low as .5 volts you'll get some fuzz but it won't be rocking very hard. If you can't hit at least 1.3v with the trimmer maxed with a particular transistor (i.e. the one from your Smallbear set), try another with less leakage.

Let me know how it works out! 

[tenken]

Thanks for your response Bret.  It helped a lot and it's interesting how the trimmer is simulating leakage...makes a lot of sense.

I finally got a chance to try out a whole bunch of transistors (OC77, AC128, AC125) at different leakages and gains.  I didn't find there was a huge difference between varying levels of leakage, although I was only really going from -1V to -2V (must have been a warmer day then when I checked them previously).  I did find differences between transistor type though.  The AC125 was a bit harsher and became a bit muddy when taming it with the tone control.  The OC77 sounded really nice, just that bit fuller. 

I can't really remember the AC128 because I had it's pins reversed for most of the testing.  I thought for some reason that the pinouts are reversed between the AC128 and AC125, so the AC128 was biasing well.  Is this correct, the pinouts on the AC125 and AC128 are the same, tab = emitter?  I've read differing things.

But yeah, there is a big difference in sound between flipping the transistors, as you'd expect.  The normal way (bias -1.5V) is very high gain, doesn't clean up well, sounds like a few videos on youtube which is good.  The reverse way (bias whereever I like) is much lower in gain and volume but can still easily hit unity.  Bit muddy on the OC77 and AC125 but sounds great on the AC128.  But it cleans up well and you can get more volume by biasing slightly higher, for example -6V.  Out of interest, do you do any damage by having the transistors reversed permently?

[Bret608]

I don't think that does any damage. Some people swear by reverse biasing to dial in a particular fuzz to their taste. My Pastyface certainly doesn't clean up with the guitar's volume knob all that well, so I found your results interesting.

I think tab=emitter on those AC transistor. That was the case on an NPN AC176 I had anyway.

Anyhow, I'm glad it was helpful and that you're liking the fuzz!