Hendrix Fuzz Face

[petemoore]

  Just throw whatever you got that you know'll 'work' in there and go gangbusters dialing in the range of available tonal characteristics the pedal is known for.

[newfish]

I have a silicon FF clone.

Q1 is 2N222A, Q2 is BC107 (no suffix).

Q1's maximum gain is 300, Q2's is 450 (minimum of 100).

...and it does clean up when the volume rolls off.

Input cap is 2.2u or 0.001u switchable in case you're interested.

[petemoore]

  Seems like there's a limit.
  Hitting the limit with <medium gain transistors...seems to hit the same limit pretty hard that higher gain transistors will, albeit in a different less forceful manner.
  High gain Fuzzface is a much different sounding circuit than lower gain.
  Because it is easy to throw the transistors in and have extreme FF Fuzz, it makes the circuit appealing for a number of reasons, one being extreme Fuzz.
  That is a fun sound if tamed a bit.
  All this is moot, there's a gain knob and other variables...
  Shoot...any Fuzzface circuit can sound alien to 'itself' under the right [wrong] conditions. that's I guess why it called a face.
  Under all the Fuzz is a Clean-ish Face, having a nice transition from using it as Clean-to Boosty-~treblish-distortion-to Thick Fuzz, available at the guitar as CW volume twist, finding the cleaned up face under all the Fuzz is a more demanding trick, [demanding in this case doesn't equate to increased desirablity necessarily, but LG transistors are worth a try if you want the ff-clean-up thing].
   The Axis Face can do it.
   Set up 'right', the tricks work great.
   A Fuzzface doesn't work alone, big into the group thing, but as such can be extremely picky about 'who [tube amp?], how [set dark?], what [guitar?]' is in the group to an extraordinary amount, I tried many random groupings with little or no preparing the group [this is how I was able to tell what was/wasn't acceptable to them], FF could produce only useless tones to catostrophic results in some scenarios.
  Since I go for the LG FF's, trying a HG FF in scenarios where FF I did try didn't work out, might have been the ticket to fierce tone which....I never tried !
     

[Renegadrian]

I was playing with my Si FF, my fried I lent it to returned it to me...
I used this layout here.


I got 2 bc108C, both have 640 as hfe...Maybe too much?! The circuit oscillate a little at max gain, you just have to turn it back a little and it's ok - anyway this thing is quite gainy!!! Maybe the trannies are too much powerful?!
Also I found it has a slight volume drop - no 500k, I used a 100k pot, as per the layout above.

[biggy boy]

I was playing with my Si FF, my fried I lent it to returned it to me...
I used this layout here.


I got 2 bc108C, both have 640 as hfe...Maybe too much?! The circuit oscillate a little at max gain, you just have to turn it back a little and it's ok - anyway this thing is quite gainy!!! Maybe the trannies are too much powerful?!
Also I found it has a slight volume drop - no 500k, I used a 100k pot, as per the layout above.

I made the same one with the 108 and find the output is low also.
So you are saying to replace the volume pot with a 500K to boost the output?

Have you tried a 2N2222 for Q1

[Renegadrian]

All the other FF schems/layouts around feature a 500k pot - I don't know if that can add a dramatic change (apart from how less you have to turn it compared to a little value pot)

Haven't got a 2n2222 around - tried a couple of 2n3904 and they seem to be more stable...
As this FF is intended to be owned by another person, I'll give him both sets...

[petemoore]

  ''Shoot for'' suggestions:
  Lower gain Q1 than Q2.
  max gain of any Q something like 200, for more fierce over the top gain, try higher Hfe's...depending on the gain of the Q's, the mannerisms of the circuit can vary a lot.
  Raise the 470R for to increase amplification output.
  500k pot lets a little more out, and when turned down some, the large series resistance introduced tends to change the EQ [cuts highs].
  Raise gain pot R [add a little series resistor or use 1k2 pot to raise gain]....set gain knob R, and adjust 470R to where it 'should' be.
 

[Gus]

Listen to the 4 sound samples you can find in the thread below from another forum.

http://www.prodigy-pro.com/diy/index.php?topic=32333.msg392319#msg392319

[liquids]

For a while I greatly disliked every 'magic' 2 knob silicon FF I tried...  To me, the circuit needs a lot of 'exaction' so that it's not too bright/dark, just the right hfe, etc, to get the exactly right amount of fuzz with the gain all but maxed, etc.  I've ordered a bunch of the magic transistors (the BC108) included, tested and tried to get in the proper hfe range, but that none of them did it for me.  I just can't stand the idea of praying I get the right hfe either.  Then, even with the mojo ingredients, I was altogether unimpressed--certainly didn't work for my rig, no gilmour or hendrix happening.

From my findings, a lot of it has to do with controlling EQ and loading.  Most FF control neither, so your left to whatever happens naturally, praying that a cap or transistor swap will get closer to waht you want.  An input impedance pot at the circuit input, to load your pickups down and hence suck out their drive and their high end with the whole circuit followed by a tone control goes a long way to making any old silicon fuzz face pretty amazing, and smoother in a FF type way, even as a humbucker guitar--->fender amp player. 

As an aside, The EZ face does sound good and is a good compromise between silicon and germanium, and it's far more consistent, though I think a tone control and impedance control to tweak high end is still helpful for tuning in the sound given the other variables.  That the only FF circuit I've tried thus far that was at all 'close' for me, stock.   The EZ face also actually has a range of FF-type fuzz available rather than just one good setting take it or leave it, unlike like classic FF and most of the the variants. It's PNP though, too, so it has that 'complication' (I hate batteries) plus it is half germanium...so I digress.

For all-silicon variants, once I discovered and tried an input impedance control, I was grinning ear to ear.  A fair percentage of the affection for the vintage Germ FF sound is how the circuit loads your pickups, but not all FF circuits--especially silicon varieties--load your pickups, and no two do it the same.  The loading --if not naturally via the 'right' hfe germanium transistors, than via pot--serves to suck out high end in a distinct way, and can likewise suddenly make the fuzz level very controllable via guitar volume, which is another typical complaint about most silicon FF variants.  With a circuit that loads down your pickups in a controlable way, you can get all of that FF goodness--even with the following 3-transistor 'high gain silicon' type of boost into-FF or tonebender variation--whatever you want to call it...but stay with me...

Try this: put a good silicon-type booster with otherwise high input impedance (i.e. you'd have to tweak an LPB) feeding a bog standard silicon FF or other variant (like Gus' Fuzz Face, YAFF, Red Fuzz, etc), using whatever silicon transistors you have.  2N3094 to MPSA18, whatever, to start.  Put a variable 'input impedance' pot before the input cap of the booster, something like 100kA or 250kA that will control how much your pickups get loaded down/ how much signal gets sucked to ground.  Following the booster stage, Put a gain pot to ground--no need for 500K audio, something small will do fine. This will control how hard the booster stage slams the fuzz face, so you to speak. You probably won't need the standard gain pot on the FF.  To start I'd hard wire it with just an emitter resistor with high gain transtors, or include the cap, but connect it right to the emitter if you want extra gain from lower gain transistors.  Add your favorite tone control at the end to control the fizz--ideally something focused around on shaving high end. I'm a big fan of following the volume pot with with a bjt buffer stage feeding an a rat-style tone control, but YMMV.  You'll obviously need to tweak caps, transistors and gain levels in all the ususal ways to your taste from this point depending on what you want, but it's pretty interesting what happens to gain/tone/volume cleanup between the gain control and the input impedance control, espcially with the latter is set low in the 4.7k range, for example.   

This has gotten me some of the best results for vintage-style FF-type lead tones that I've ever tried, hot bridge humbuckers and fender amps included!  But it's not for a FF purist.     I'm definitely still tinkering with this type of circuit, but it's killer and worth trying....  If that is too vague a starting point for to follow, PM me for a schematic example along those lines for clarity.   

And no matter what circuit you end up liking most, don't forget a little delay/echo after the fuzz for to for extra vintage psychedelic mojo.   

[Gus]

The feedback resistor does a few things,
Helps set the closed loop gain around the first transistor this helps set the interaction with the guitar volume control.
Adjusts the input summing node, more feedback and the higher the gain of the transistor the closer to the ideal 0 ohm summing node
Has some affect on base current for the first transistor, think pinch control

Look at the Hot Si in the Gallery.  The first stage has designed input resistance(pickup loading) and output resistance(works with the 47K feedback resistor as part of the next stage gain setting) and a gain of about X10.

[liquids]

Look at the Hot Si in the Gallery.  The first stage has designed input resistance(pickup loading) and output resistance(works with the 47K feedback resistor as part of the next stage gain setting) and a gain of about X10.

Now that makes sense...thanks.  While I'm not as wowed by the tone of the Hot Silicon sound clips have never struck me as all that fuzzy, sound clips are always highly variable.  Still, that's an oddly similar starting point to the ideas I was getting at, conceptually and schematic wise!--though, I never paid attention to that circuit previously, due to the sound clip. 

Q: Q4--since the signal off Q3 is pretty hot, does it contribute some "fuzz" of it's own, or does it end up biasing correctly without clipping the rails?

While I prefer the ability to dial in the amount of loading, R1 and R2 could also easily be swapped for 33k/4.7k, 22k/3.3k, or 15k/2.2k to get in the range of maximum loading the user prefers... Likewise, one could easily just replace C4 for a bigger cap like 22uF, and add a gain pot between Q1 and Q2 for more control over gain.

[col]

Don't worry about the silicon transistor's number. Use some sockets and see what sounds best. I found with two BC109s, BC108s, BC107s etc. the initial fuzz sound was fine but the decay was awful. IMO low power transistors sound much better, BD139 power transistors are fine (different pin out, BCE rather than CBE) as are the free ones in broken low energy light bulbs. I also got some very low hfe Si transistors of 45 -55 which compare very well with my matched pair of Ge FF fuzz. Good initial fuzz sound  and smooth sustain into musical feedback. Beats the pants of an angry wasp metal pedal type sound. The PN2222s I tried were quite good which are a 2N2222 with a plastic body. Lots of variations with this basic circuit  (VOX distortion booster, stoner fuzz etc). Try as many as possible, you won't believe the difference in sound for a slight change in component values.

[Renegadrian]

Right, I believe that as for the TS you have two ways...Build it as per the schem, with minor adjustments if needed, or go with severe mods, till there's not so much left of the original...

We know there are countless FF reinterpretations, but what if you want to stay close to the original?!

[Brymus]

I like a 2n3904 for Q1 and a 2n5088 for Q2 the bias is different with every set.
as low a 3v as high as 6v but thats why you use trimmers and sockets.
I get good Hendrix tones with my set up and thats about 75% of my needs for a FF.

[Renegadrian]

Yeah I believe that a couple of 2n3904 are a good solution - The ones I got are in the 230 hfe range and sound good - I guess that the 640 of the other set is just way too high...

[solderman]

Hi
Well I think that 2N2222A is the best option for both, but then again it comes down to what sound you like. Since the Fuzz Face design is sensitive to where you put it in the pedal chain that's gonna affect the sound as well. I think that direct after the guitar gives a good Hendrix sound. With something in between the heights is going to be too pronounced.

This is a soundsample to compare two Fuzz Face's I have made, one Ge (the firs in the all tree soundsamples) and a Si (2N2222A)

http://solderman.fatabur.se/Future%20project/Ge-Si%20_shoot_out.mp3

[liquids]

I should now also say (if I'm allowed) that the "Tweak Fuzz" is a simple silicon FF variation I tried recently, and it is easily the best sounding 'standard' silicon FF I've tried thus far--if you want a good sound, and a circuit that is more similar to the original circuit, try it.

[Derringer]

Don't forget about piggybacking.

I'm playing around with another piggybacked tonebender design now. Right now in the Q1 position of the FF section, I have a pair of 2n5089 that have their bases tied together along with a 50K trim between the emitters (20K would probably give some more tweakability). I started with a 2n3903 with hfe ~ 120 in the Q2 position of the FF section, but then tried another 2n5089 and like that sound better ... more crisp.

Tweaking the 50K trim offers some really nice variability in breakup. Every time you alter the gain of the Q1 section though ... you must compensate the bias on Q2 .... so have a trim on the Q2 collector as well.


The feedback resistor, as Gus has pointed out many times before, also plays a very big role in the overall character of the sound. Running small caps parallel to the FB resistor do alot for EQ too. When the FB resistor gets adjusted, again, be sure to compensate the bias of Q2.

[Brymus]

Since posting in this thread I breaded up Gus' Rocket I didnt like it with all 5088s but using 3904 for Q1 an Q2 yielded a very nice
Hendrix fuzz with a more Metal tone to it.
Its noisy but the tone is awsome.
I would try the Rocket if your considering a FF.

[Der Groovemeister]

+1 on the piggybacking!!!
I tried a lot of mods to try and make a SI FF sound as smooth as a GE, but none came as close as actually mimicking the exact (low) hfe values by piggybacking 2 pairs of 3904's. I put a trimpot on and measured them in my DMM while turning the trimpot. I made Q1 80 and Q2 100. I also put a 3n3 cap in parallel with the 4k7 res. like in the Axis Face to cut some high. That's pretty darn close to the GE sound! I think it sounds even better than the already great sounding Axis face. And with the hot weather we had here last week, my GE FF totally farted out but my piggybacked SI still sounded super! 
I also put in a pre gain pot and left the normal gain pot out (who uses that one?!)