Need help with Fuzz Face transistors

[thermionix]

Just joined, first post.

Background:  I build, repair, modify tube amps.  20 years.  Self-taught.  I know very little about solid state stuff, but I can solder like a mofo, follow schematics, etc.  I've built a couple BYOC kits, and a couple FFs from scratch over the years.

A few days ago I decided to have another go at a FF, and this time tweak on it until I actually liked the sound.  Been doing research here and elsewhere online.

Read and re-read RG Keen's geofex FF page and decided to build the leakage/gain transistor testing circuit he outlines there.  My Fluke doesn't have a transistor test function, so this seemed ideal.

Whipped that up this evening.  My resistors measure 2.200M (exact) and 2.488K (surely close enough).  Brand new 9V battery.

I have some AC128s coming in the mail, but wanted to try out the tester, and had 2 new/NOS NTE126As laying around.  Tried 'em both.  Getting strange results.

Both show virtually zero leakage, and VERY low gain, 6 and 9.  I was expecting results closer to 40 or so.

Is this normal?  Does this tester only work for certain transistor types?  Are NTE126As really that low gain?  I've read somewhere that they work in a FF, even though maybe they don't sound particularly good.

Any help is appreciated!  Thanks!

[MaxPower]

The 126s are supposed to have gains of about 40. Double check that you have your polarities right: collector connected to ground/-9v, and the emitter connected to the positive rail/+9v. Assuming the 126s are PNP transistors anyway.

Edit:
D'oh! You've built ffs before so you probably have your polarities right. In that case, I don't know.

[mth5044]

Is it possible you lost a decimal place somewhere? 60-90 hfe rather than 6-9?

[thermionix]

To say that I've quadruple-checked everything would be an extreme understatement!

Yes, NTE126As are PNPs.  I got the pinouts from NTE's datasheet, so that should be correct.

For the gain test I read .06 volts across the 2.4k resistor.  Multiply x100 for a gain of six.  Isn't this correct?

I suppose it's possible both these transistors are duds, but I find that unlikely considering the fact that they're not used.  I don't know where I got them, they've been in my "random solid state parts" drawer for years.  They're the only PNPs I have on hand until my AC128s arrive.

[Ben Lyman]

I have one NTE126 that I was going to make a Rangemaster with... no go... almost zero leakage and HFE of 12. I don't buy NTE stuff anymore. Maybe it's hit and miss, some work/some don't, I don't know.

[thermionix]

Maybe that's it.  Will know more when the 128s show up, I reckon.

[duck_arse]

just for chuckles, stick those NTE's in 'backwards', and see if the gain increases.

I did this same RG-ing on a small batch of russian Ge's, had a hell of a time getting the pinout to agree w/ the real world, even with the datasheet. measured them all backwards, got better results the other way.

[thermionix]

What is backwards?  Flip C and E?

[duck_arse]

in my case, yes, that's what I did. well, coming from the other direction, but yes.

I thought something was wrong when I started to get hFE's of "1".

[thermionix]

I tried that last night.  Leakage test gave me ~2.4v across the resistor (1mA leakage!), hit the gain test switch and the voltage stayed pretty much the same.

Well, that can't be right.  Also I trust NTE has the correct pinout for their own product.  Must be duds.  Not shorted, I checked for that.

AC128s should come today or tomorrow.  We'll see what happens.  i ordered 4 to pick the best 2.  I was hoping the NTE126As would give me more options, but apparently not.  Oh well.  I don't think I actually paid anything for them, they just...gathered in my orbit somehow...you know it happens.

[kaycee]

Sometimes transistors act different in the real world to how they measure on various test equipment. The Fuzz Face is a simple circuit to breadboard, or even just knock up on a scrap of vero or perf with sockets. Then just plug the transistors in and see what's what.

If your NTE's are low gain you'll get weak fuzz, doesn't mean you have to junk them, read up on piggy backing and you may yet find a way to use them.

[thermionix]

Yeah, dig.  I'm gonna use sockets, so I can try various combinations.  If/when I come up with a setup I really like, having the test numbers will make it more repeatable, if nothing else.  Also, I'm just trying to learn about this stuff in general.

I really don't use pedals too much.  Rarely more than one at a time.  I haven't played very many FFs, and the only one I liked was quite low gain, and much more midrangey than others.  Sounded almost like a tube screamer.  It was also the only vintage one I've ever played.  It was either silver or blue, don't remember.  Couldn't tell you which transistors, but one had been replaced.

Not that I'm trying to recreate that.  If I can get a lot of gain, good-sounding gain, then I could dial it back if I wanted it lower.  There's a knob for that!

I'm really looking forward to this project, should have final components Monday or Tuesday.  Pretty sure I'm gonna add the Fuller-style input pot.  Haven't decided yet on a bias trimmer.  Leaning toward lowering the input cap value.  Options...I got 'em!

[Ben Lyman]

If you follow the Small Bear "breadboarding a silicon fuzz face" article step by step (like I did 9 times) you will eventually reach the step where you can make it sound however you want. That sweet midrange plus crunchier, lower gain  happened for me with a slight twist of the 10k bias trimmer. You can also try different caps on your breadboard to alter the tone a bit, or even put two caps in parallel with a switch... or a blend pot... or... or... 

[thermionix]

Yeah, it's gonna be fun.

I don't have a breadboard, but the circuit will be built on an eyelet board, so I can sub components 'til the cows come home.

I've got a metric shit ton of vintage caps, and always keep a full stock of half-watt carbon comp resistors on hand for amp work.

I WILL dial in something cool!

[thermionix]

Well the AC128s came today.  Definitely getting different results!

Can someone familiar with this test PLEASE double check my math/process?  Maybe I'm misunderstanding Mr. Keen's instructions.

Example:

For the leakage test I read .200v across the 2.4k resistor, for a leakage of about 80uA.  Is that right?  For the gain test I read .440v across the same resistor, for a net gain of...negative 36 (44-80)

What am I doing wrong?

[Electric Warrior]

it's 44-20. 

[thermionix]

Finally, I see it.  I was subtracting the wrong damn number!  Thanks!

HFE of 24 sounds pretty low, though.  But those numbers were from the weakest one of the four I got.

GT-AC128s from Mammoth.  What are people's experiences with those?

[duck_arse]

also, the leakage is 200uA, from how I read the instructions. I can't say for mammoth, but against italianguy63's stock, 200uA is quite low. also, hFE24 is quite low, my batch of four read 58 low~70 high (sydney summer readings) and leak 500~650uA.

as an aside, I was recently working on a fuzz, trying diff Ge's, from japanese, chinese, russian, u.s. and AC series. every time I swapped in an AC128 or AC188, I was surprised by the immediate tonal change to "DOOM".

[thermionix]

Not to argue too much, I'm the newb here, but from the way I read it (not to mention Ohm's law) you divide the voltage drop by the resistance to calculate the leakage current.

What do you mean by "DOOM?"  Bad sound, or massive distortion...or what?

[duck_arse]

hmmm, and as I often say, "MY EYES!" yes, I now re-re-re-read the RG page, and see all my measures are wrong for leakage, probably forever. oh well, sorry about that, everyone.

when I say doom, I mean glum, gloomy, downcast. not exactly dull, I think some might say "dark", but probably a few shades darker than that. not bad sound, not more distortion, just a change of tone and a small sample size.