Fuzz Face Biasing Problem

[Wehrdo]

I'm building my first pedal, a Fuzz Face, and I'm having trouble getting the right sound.

My transistors are a matched pair from Small Bear Electronics, with a reported gain of 68 and 136. When using the resistor values that came with it (33k, 680, 11k, 120k) there was HUGE gating, and the collector of Q2 was at 0.7V. When I adjusted the resistor values to get a decent fuzz, they were a little low low(33k, 470, 5.2k, 100k) and Q2 collector was ~2.2V.

When setting the values so Q2 is 4.5V as recommended, my biasing resistor is really low(2.2k). It also has a very bland sound. Here's a recording of first straight from the (bass) guitar, then with the pedal. This is with the tone knob all the way up.

What's the deal? Why does the biasing resistor have to be that far off, and why isn't there any character when it is? Thanks for the help!

[rousejeremy]

Check the pinout of your transistors. I bet you have one backwards.

[Wehrdo]

Assuming the have a pinout just like the third picture here:


then they are in right. I've actually accidentally put one in backwards a couple times while debugging, and it made no sound at all. It's on a breadboard.

[LucifersTrip]

what are the full voltages of both transistors?  what are the transistor part numbers?

[Wehrdo]

With the standard resistor values (470, 33k, 8.2k, 100k), the voltages are as follows:

Q1:
E: 0;      B: -0.76; C: -1.1

Q2:
E: -0.95 B: -1.1    C: -0.98

It also sounds hideous with those values.

[EATyourGuitar]

looks like Q2 BC is reverse biased. this does not mean it is definitely backwards, it just means you dont have voltages c < b < e (pnp). Q2 E and Q2 C are very close so maybe they are connected on the breadboard somewhere they should not be. you could also try and flip the transistor around and test voltages for Q2 again and post that.

[chi_boy]

Are you positive you have the resistors in the correct spot?  I ask because the first FF I built with Steve's transistors did the same thing, but I had the wrong values in the wrong place.  Steve uses the resistor numbers from his own Bear Face schematic.  You need to be sure yours are the same.   

[rousejeremy]

There's a breadboard layout of the Fuzz Face here if you'd like to try it out. http://www.beavisaudio.com/bboard/projects/

[Wehrdo]

Well, I rebuilt it using this breadboard layout, obviously switching the polarities because I'm using PNP transistors. Some voltages, same sound, a lot more jumper connections, and looks messier. 

[smallbearelec]

This is a PNP All-American pair, so the transistors should both have 2Nxxxx designations. Pls let us know if they don't, because that would be a mistake on our part. To review, according to the resistor values that we spec'd:

Q1 collector load should be 33k
Q2 collector load should be 11K
Q2 divider resistor should be 680 ohms
Feedback resistor Q2 Emitter to Q1 Base should be 120K

If it is wired right and with these resistor values, the Collector of Q2 should hit about 4.5 Volts. If you can't figure out what's off on your own, please drop an e-msj to the smallbearelec address and we'll try to help you out.

[Thecomedian]

I have a theory that has generally held true for my transistors, which is that if you have the Positive and Negative leads touching the proper transistor pins, you should get a higher resistance through vBE than through vBC, set to ohms. tested outside of circuit of course.

here's some PNP version schematics, if it'll help.

http://www.muzique.com/lab/fuzzface.htm

Like those other links talk about, changing the resistance on the Collector of Q2 can have good results. I use a variable resistor and twist it with a screwdriver to find the "sweet spot", and this can vary from being 5k to being 22k. Inbetween it usually sounds limp, and that's just ONE resistor value to mess with.

That buzzing sound sounds like bad ground or a reversed transistor..

[Wehrdo]

My transistors are Motorola part number 2N508 (7313).

Tested with negative lead on base:
Q2 BE resistance: 1.129k
Q2 BC resistance: 1.160k

Q1 BE resistance: 3.83k
Q1 BC resistance: 3.96k

Opposite of what you hypothesized, so that's not good.

And the buzzing was only without the pedal. It's just bad recording equipment; i.e. crappy laptop

[petemoore]

  HFe Checker [in DMM...shows expectable gain # in the readout when the pins match pinout, note how the pins are marked when it's making gain in the meter.
  Alterante transistors [mainly if you've got sockets], something low gain or plain 'ol 2n3906 [PNP] should do, they're sturdy, have very predicable/predictables and pinout is very easy to check/triple check.
  anyway that's how I used to do it

[Wehrdo]

Unfortunately, I don't have a DMM with a transistor checker, nor another PNP transistor, and it would be several days before I could pick up one since the nearest RadioShack is in another town.

But I'm certain the middle lead is the base, because I only get resistance readings when the negative lead is on it. And if the emitter or collector were switched from what I thought they were, I wouldn't get any sound. Also, the datasheet says it's a TO-5 package, which has pinouts like I posted above.

[LucifersTrip]

the small tab is above the emitter...and here's a simple hfe checker you can build

http://www.geofex.com/Article_Folders/ffselect.htm

[Wehrdo]

Oh I'm an idiot... They were backwards the whole time.  Please forgive me.

I swear I did turn them around when it was first suggested, but I was using some crappy sockets, and there could have been a disconnection.

So, now I've got it working, another question. I feel the sound is quite harsh. When the fuzz knob is turned all the way down it's at a nice distortion level. Anything much else and it's unbearable. What sort of mods can I do to make it less harsh?

Thanks for helping me through all this, I really appreciate it!

[smallbearelec]

Please forgive me.

Nothing to forgive...such mistakes are a normal part of learning. Now that the thing works, look at the voltages and the bias Again. The circuit should not be at maximum fuzz with the pot all the way down IF all is wired exactly right AND the bias resistors are what we initally spec'd.

[Wehrdo]

So here's how it sounds, using suggested resistor values. First at lowest setting, then middle, then highest. Kinda harsh for my liking.

This video has the sound like I'd like to get. In the comments he says it isn't modified.

Oh, and it's biased at 4.5V.

[Jussi]

Sounds like the transistors aren't biased right.

[Wehrdo]

I've added a 33pF capacitor between the base and collector of Q2 to soften the sound, and changed the input capacitor to 4.7uF and the output to 0.047uF to increase bass response. It's sounding much better now.