some Fuzz Face questions and problem

[Speeddemon]

Hey guys, I made a Fuzz Face, but with the 100k log volume pot, instead of a 500k log.
I used 2N3904 transistors (Si NPN, hFE 250-280), but now with the fuzz control open, it squeals in a very high pitched tone.
If I back off the fuzz pot only 10% or so, it's gone.

Does that have to do with the 100k pot? I read somewhere it adds treble. Was it more intended for low-gain Ge transistors?
Is there a way to eliminate this squealing without changing the pot? 


And what effect does changing the 1k lineair Fuzz pot to a 2k pot have?

[BrianJ]

It probably has nothing to do with the pot I use a 50k in mine.  Whats happening is the frequency response is too good and the pedal oscillates at maximum gain.  It may have to do with component layout, parts selection, wire paths... a good fix may be to parrallel the 33k collector resistor with a small cap (.001uF-.01uF).  The bigger the cap, the more treble you will lose, so adjust to what is necassary for stability.  Sometime this aproach will cause or "tune" oscillations, but its a good start if you dont have much troubleshooting experience.  A grounded enclosure may help as well. 

[Speeddemon]

It probably has nothing to do with the pot I use a 50k in mine.  Whats happening is the frequency response is too good and the pedal oscillates at maximum gain.  It may have to do with component layout, parts selection, wire paths... a good fix may be to parrallel the 33k collector resistor with a small cap (.001uF-.01uF).  The bigger the cap, the more treble you will lose, so adjust to what is necassary for stability.  Sometime this aproach will cause or "tune" oscillations, but its a good start if you dont have much troubleshooting experience.  A grounded enclosure may help as well. 

Thanks! I'll try that. The weird thing is just that I built a similar one, with the 500k pot, and that one had no problems at all (and the same PCB layout (& same trace-width) too!)
I do always ground the enclosure, by using the open Switchcraft jacks and a metal washer (I used plastic washers before, but they caused hum sometimes)

[Sir H C]

Hey guys, I made a Fuzz Face, but with the 100k log volume pot, instead of a 500k log.
I used 2N3904 transistors (Si NPN, hFE 250-280), but now with the fuzz control open, it squeals in a very high pitched tone.
If I back off the fuzz pot only 10% or so, it's gone.

Does that have to do with the 100k pot? I read somewhere it adds treble. Was it more intended for low-gain Ge transistors?
Is there a way to eliminate this squealing without changing the pot? 


And what effect does changing the 1k lineair Fuzz pot to a 2k pot have?

the fuzz pot value will change the bias point unless the collector resistor is doubled also.  What happens depends on the transistor, but you are doubling the output impedance of the circuit with the 2k so then you lose more gain when you go to the output.

[petemoore]

  Once it's 'working', it's a matter of tuning it in to do what you want it to.
  Turning the gain down or trimming the max gain so it never ocsillates is one solution. [see Geo secret life of pots, try a 18k across lugs 1 and 3 of pot, experiment with values, measure with DMM paralleled R's value} But..
  I'd recommend leaving the circuit mainly untouched, if you have a Q2 collector pot and have socketted the transistors, and 'interview' some different lower gain transistors in the transistor sockets, rebias/redial/retune/revoice? check bias at least after each swap. This will exhibit the range of gain levels and frequency responses the circuit is capable of, a comparitively wide range.
  I recommended to myself a FF testboard, like a breadboard, but more permanent. I used perfboard, socketted or trimpotted most of what's there, for instance the 470ohm, put in as 1k2 with 'longleg' above the board, I can clip or tack a parallel resistance there to see what I like there.
  There are a million things you can do with a FF, lots of work has been done to the 2Q, FBLoop Amp circuit from which explanations abound
  Drive yourself nuts with the re-tweeks...or...
  Check out Axis Face 3 [Fuzzcentral],
  5yr FF Vet, is about to rewire 'YADFF' to SUPAFACE, in Dragonfly's Gallery Louge, which BTW is a great place to hang out, check mods.
  Joe Davissons Analog Anchemy "Fuzz Face Calculator.
  GEO Technology of the Fuzz Face

[Speeddemon]

I experimented with different transistors already:

2x 2N3904= singing, soaring high gain. Lots of sustain.
2x BC109C= soaring/fizzy high gain. Not really harsh, but more raw/gritty than 2N3904.
2x BC107A= super bright, harsh screetching high gain. NOT recommended!
2x 2N2369=round, singing, medium gain. Mediun sustain.
2N3114+2N2369= even rounder, low/medium gain.

[Sir H C]

Have you tried any supply decoupling?  Cap across the rails might fix the problem.  Sometime too if you split the first stage collector resistor and put in a cap, that can help too.

[Speeddemon]

Yeah, there's a 100 uF from +9V to ground for decoupling.
The squealing seemed less after I added (as per advice of R.G. Keen's Fuzz Face technology article) 100pF caps from collector to base on each transistor.

Still, it's weird that the similar built one didn't have that problem.

[petemoore]

  FF's 

[brett]

Hi.

Still, it's weird that the similar built one didn't have that problem.

Because the FF runs the first transistor flat out, and the second one very hard, things like transistor-to-transistor and capacitor to capacitor variation (same type, even same batch) can lead to differences.  For example, your 2N3904s probably have hFE in the region of 400, and any two might differ by 20 or more.  Also, the 22uF cap probably has up to 40% tolerance in the value, which can affect the gain through Q2 when a FF is cranked to 90% or more.

cheers

[b_rogers]

i seem to have fixed this problem before with a 1k-5k resistor in series just after the input cap,then a smallish 470pf or so cap to ground in between the incap and q1's base (input).

hope this helps..it should help if you are getting any radio stations too


Brent

[vanessa]

Are you using a 330R (or 430R) rail resistor or a 1K2 resistor like on the fuzz central article?

I've found that this 1k2 mod not so pleasent with high gain si transistors and certain germaniums. Issues very similar to yours with this 1k2 and most higher values. I would stick to the original resistor values, it's helped me.