EKI FUZZ

[Dragonfly]

Found this on ebay...the pictures are tough to read, but I think the schematic i've redrawn is correct, though I made a couple changes.....i added a master volume, and eliminated a 10k resistor to ground ...

Let me know if you breadboard / build it...

If you see any mistakes, let me know...

AC

[soulsonic]

Interesting. What is this from originally? It looks like the instructions from a kit.

[Dragonfly]

Interesting. What is this from originally? It looks like the instructions from a kit.

Yep...its a kit fuzz thats on ebay right now...

[petemoore]

  Thanks for posting, Dragonfly, yet another Fuzzer ! Different looking than the other ones !
  Large Q1CR and Feedback resistor, also 50k is large for an emiiter R..wonder what it sounds like.

[Dragonfly]

  ..wonder what it sounds like.

Probably horrible.

[MartyMart]

I have a schem of a fuzz that's 99% the same and it sounds quite decent !
First r off Q1's emitter is 1k and the fuzz pot is a 10k though.
I changed the input cap to a 22n and that sounded great.
MM

[brett]

Hi
that "fuzz" pot does 2 separate things: set the AC gain of Q2, and apply feedback to Q1 (reducing gain and increasing the input impedance).  So there might be quite a difference between using a 10k and a 50k pot there.

I'd be tempted to "split" the 50k into a 10k pot with the fuzz cap on its wiper, and a 22k resistor (try 10k to 100k, larger values = less feedback and more gain) in series with the pot, leading back to the emitter of Q1.

Also, for creamier tone, you can parallel the 6k8 resistor on the collector of Q2 with a 4n7 (0.0047uF) cap.\

cheers

[slacker]

cool, it looks like this one http://www.muzique.com/news/index.php/?p=33#more-33 but with the switching removed.
Might be worth a build.

[Steben]

My instinct tells me clean notes are harder to achieve with such high feedback resistor.
the lower this value, the higher the base bias at Q1 and thus the more headroom.

[petemoore]

  yupp yupp..
  50k emitter R value there is like wayy bigg for that.
  4m7..seems like it'd sound about like if it wasn't there.
  680k Q1CR...
  Definitely looks like a 'pinch-bias' fuzz, which might not let smaller signals might not 'push' through.

[Gus]

The .001uf does not go to ground but to the node of R7 and R8 looking at the scanned schematic. 

[Mark Hammer]

  Definitely looks like a 'pinch-bias' fuzz, which might not let smaller signals might not 'push' through.

And, while not stated quite so technically, I think that was one of the chief complaints we had about fuzzes in the 1960's; they simply couldn't achieve subtlety.  Or at least a lot of the ones my friends and I encountered were like that.  Either you picked hard and had an obvious fuzz, or else things seemed squelched.

[Dragonfly]

The .001uf does not go to ground but to the node of R7 and R8 looking at the scanned schematic. 

.001 goes to R8 and then to ground, but as I explained in the first post I eliminated R8 (10k)....

Do you see any real reason it should be there ? Just wanting your opinion.....

I guess as I look closer. the .001 also (as well as going to ground through R8) goes to output....

I'll wait till it gets breadboarded, but I might make that change in the schematic...

Thanks Gus.

[Gus]

Maybe one should just build it like it is scanned or connect the .001uf to the wiper of the pot I would use Si transistors with a  hfe around 200 as a first guess.

Now there is a few things going on in that circuit.  Q1 is a gain stage the collector can be thought of as the emitter of Q2 Vbe drop away.  DC bias for the first stage is from the 4.7meg the max advailbe current for Q1 base is set by the operation points (50K pot and 470 ohm Q1 emitter resistors voltage drop) and the 4.7meg resistor Q1(it does look like it might be a pinched fuzz as noted by Petemoore).  Also there is a feedback loop kind of like an inverting opamp/fftype circuit from Q2 e to Q1 b 4.7meg and 47K.  Note how the emitter currents are summed together in the 470 ohm.  One transistor is turning on and other off.  Q1 has a large value collector R compaired to the emitter R.  Q 2 has a small value collector R compaired to the 50K pot

The emitter of Q2 is inverted compaired to the emitter of Q1 the .001 connects to the output node of R7and 8 the 50k wiper cap passes higher frequences to the output kind of like a blend pot.

A lot going on in what looks like a simple circuit

[Dragonfly]

For those wishing to build the original, unmodified version, here's the redrawn schematic. It should be correct.