High gain PNP in Tone Bender?

[Greenwichpaul]

I reworked my 80s Fuzz Face into a monstrous beast with info from this site, but it started an addiction. Since then I've acquired some nice enclosures and a bag of Siemens AC125, from a tiny shop that sells lots of NOS.

I really wanted to make a Tonebender 2, as a contrast to the Fuzz Face, and maybe a 3 as well.

But I did a quick test with my meter, while I wait for the right resistors to arrive for the proper gain/leakage test. And everything is gain 120-200. My experience is that will come down circa 10 per cent but not a lot.

Is a gain 140 useable in any position? Has anyone used them, or 180 or so, in any circuit?  Might they work all though, or in combination with another type?

I'll breadboard them up but would like some clues  because I'm very indecisive and could spend years of my life deciding which sounds best. I don't mind buying a few more lower gain PNP but would like to use these if I can. They were only around 50p each but I'd prefer to use them somehow.

TIA for any stories of your failures or successes with higher gain PNP.

[WormBoy]

are these numbers leakage-corrected gain?

[Greenwichpaul]

are these numbers leakage-corrected gain?

No, this was a quick multimeter test while I wait for my 2.472k resistors to arrive. Last batch I checked showed that, providing leakage was right, gain was circa 10% less than on the multimeter tester. I'm pretty certain these will nearly all come in high.

[WormBoy]

Would be good to test them in detail, as the Mk II would probably like some leakage, especially for Q1. Sorry that I can't help you with a success/failure story for relatively high-gain transistors in the Mk II, since I am usually in the opposite position (I am usually stuck with low hfe ones)  . However, there's no harm in trying them, since you're planning to breadboard anyway.

[Greenwichpaul]

thanks Wormboy.

Turns out these AC125 are pretty different to all the other PNP I've tried. Even more variation than, say, my CV7001 with heat plus way more leakage.

I finally soldered together RG Keen's little circuit so I could test them properly and allow them to cool down without handling . Ended up getting very consistent gains of 100-150. Average leakage was around 300μA, minimum 190, all the way up to 650 or 700. I know from the mega freestompboxes thread that some folks have got great results with high leakage versions, so it looks like I have a full selection of 20 to choose from. Despite my worries, this turns out to be the most consistent batch I've come across, hopefully fertile grounds for a Tonebender II and who knows what else.

[WormBoy]

Yeah, that's a very handy little circuit! Looks you found yourself a nice batch for making fuzzes. The 'recipes' I trust most for the MkII and MkIII advise a gain of somewhat less than 100 for Q1 and Q2, but I would just use some from the lower end of your gain range and see how they sound. The more leaky ones may be good to take a stab at a MkI  . Happy breadboarding!

[Greenwichpaul]

I tweaked the resistor to 2.47k exactly and only then realised how irrelevant it was! A difference in sunlight easily halves Ic from 1000 to 500uA.

All the TB IIs `I've read about seem to have leakage of 300 minimum, but with a wide variety of hFE from 85/60/80 and 96/135/68 (Electric Warrior) to 174/208/194 (DAM's The Wife), I thought I'd seen a pattern but providing you have that leakage of 300 plus, presumably in Q1, looks like lots of options, and a good number with 500uA Ic .

I wonder now about all the CV7001 I threw out because they had 500uA leakage...

[Electric Warrior]

Those reading are what I had in a clone at the time. I'd start out with matching gains and swap one transistor by one, listening for the difference it makes and then decide if it's an improvement or not. Just use your ears.

Generally, high gains can work and and bias just fine. The OC75 version of the MKII is biased for transistors with a fair bit of leakage.  As with every transistor, look at the voltages. If you have too much or too little leakage you can see it right away.

Measuring the hfes and leakages will help getting better control over what you're doing, but you have to actually try them in circuit to see how they bias.

[Greenwichpaul]

It's a 'how long is a piece of string' question, but if you had triplets of 80, 90, 100, 110, 120, plus 130/140/170, all around 300 leakage, what region of gain would you experiment with first?