"Choked" sound and very low volume, my first build the YAFF pedal

[Woolly Mother Mammoth]

Hello everyone!

I am building my very first NPN fuzz pedal, the "Yet Another Fuzz Face". Unfortunately I have run into some small issues (hopefully).

When the pedal is in bypass everything is fine, standard volume and tone but when I engage it the volume is very low and I have to play the strings really hard to get any kind of fuzz out of it. It sounds very "choked", buzzy and have no sustain what so ever. I figured that it could be trouble with the bias settings but I can not seem to get it right. I tried two different trannys (MPSA06 & PN2222A) and the same thing occurs with both.




Here are goes the instructions for the YAFF: http://www.muzique.com/lab/yaff.htm

YAFF original schematic: http://www.muzique.com/lab/yaff.gif

PN2222A Datasheet:http://www.fairchildsemi.com/ds/PN/PN2222A.pdf

This is what mine looks like (I had to take a B220k pot for the gain control in lack of others).

[deadastronaut]

try swapping the 6.2k for a 10-20k trimpot..and tweak...

[amptramp]

First step: check to make sure the electrolytics and the transistors are installed in the right direction with the right pinout.

If you still have insufficient output, R2 can be made larger.  You currently have a gain of about 0.05 from the collector to the junction of R2 and R3.  The original fuzz face did not have an emitter resistor for Q1 - the emitter was grounded.  The emitter resistor tends to reduce gain and reduce the distortion which gives you the sustain.

[seedlings]

Um, that 220K will kill your signal, unless you adjust it with a meter to read 500~1500 ohms (which will be pretty hard to do).  Can you put a 1K or 1200 or 1500 ohm resistor across the 220K pot?

CHAD

[Woolly Mother Mammoth]

Wow, nice response. Thanks everyone!  

try swapping the 6.2k for a 10-20k trimpot..and tweak...

Ok why not, I'll try.

First step: check to make sure the electrolytics and the transistors are installed in the right direction with the right pinout.

If you still have insufficient output, R2 can be made larger.  You currently have a gain of about 0.05 from the collector to the junction of R2 and R3.  The original fuzz face did not have an emitter resistor for Q1 - the emitter was grounded.  The emitter resistor tends to reduce gain and reduce the distortion which gives you the sustain.

Amp you little tramp! You nailed it!
The trannys was placed in reversed, newbie mistake I guess. Now I got fuzz!!

I'm definitely gonna try changing the resistors, thanks for the tip!

Um, that 220K will kill your signal, unless you adjust it with a meter to read 500~1500 ohms (which will be pretty hard to do).  Can you put a 1K or 1200 or 1500 ohm resistor across the 220K pot?

CHAD

I does not kill my signal BUT when I dial the gain and the volume pot to max, I get some kind of weak feedback loop, a constant "popping" beat.



So, I now get fuzz out of it but with a feedback loop and the clean signal mixed with it. Now that, ladies and gentlemen, is somewhat strange if you ask me. I got to look into what is causing this. Anyone got any ideas?

[Woolly Mother Mammoth]

ok, I managed to fix the feedback loop and clean signal that was interfering with the fuzz. I must have been a bit too tired when soldering the ground/output/input wires to the DPDT...  

One thing still bites me though, I tried adding a higher resistor at R2, but with out luck in the volume increase, nothing happens it sound exactly the same.
I tried bypassing R1, R2, R3 and C3 and just wired the collectors of Q1 and Q2 directly to the volume pot, that increased the volume significantly but I'm not sure it sounds all that good.

[petemoore]

  To give more to go on use the Debugging thread information, voltage measurements of every transistor pin from ground...everything else.
   Another approach is to count the # of connections at node 1, noting polarity of any polarized parts [use NPN Neg gnd or PNP pos gnd schematic for the kind of transistors you have], the measure them at the board, that every connection but no extras etcetera...
  They can be funny to bias anyway, having Q2Collector bias resistor adjustable between say 6k8 and 16k8 makes trimming the R a snap and more precise [10k trimpot + 6k8 fixed resistor adds to between 6k8 and 16k8...

[Woolly Mother Mammoth]

 To give more to go on use the Debugging thread information, voltage measurements of every transistor pin from ground...everything else.
  Another approach is to count the # of connections at node 1, notinag polrity of any polarized parts [use NPN Neg gnd or PNP pos gnd schematic for the kind of transistors you have], the measure them at the board, that every connection but no extras etcetera...
 They can be funny to bias anyway, having Q2Collector bias resistor adjustable between say 6k8 and 16k8 makes trimming the R a snap and more precise [10k trimpot + 6k8 fixed resistor adds to between 6k8 and 16k8...

I will check the Debugging thread, I should have thought of that from the beginning, sorry.
Unfortunately I'm using an old 80's analog meter not a fancy digital and I'm a total newbie at this so bare with me if something does not add up or if I failed to measure these ones right.


With the hFE meter at X10 (15mA) I get the following measures:

Q1
C: 1
B: 8.4 (I had to set the meter at "XK 150uA" to get any reading at all).
E: 10

Q2
C: 3.2
B: 1
E: 3.1

I just learned that the signal cap (C1 2.2uf) was too small? = choked signal/sound.

[Woolly Mother Mammoth]

I placed a 470uF(!!) as signal cap and now I get sustained fuzz, not snappy gated like before. It seems really strange that the "setup" needs such high cap to get the signal through.. but I guess that is what it takes.