adding a gain stage to a fuzz face circuit -- what input cap to use?

[mordechai]

I want to add a PNP silicon transistor gain stage to a fuzz face circuit to drive the gain up quite a bit similar to what ZVex does on the Fuzz Factory (without all the admittedly awesome bells and whistles on that circuit!) and I noticed that Vex chose a pretty large input cap -- 10uF, I think.  This is so much bigger than the input cap on a regular Fuzz Face, and I imagine this has something to do with providing a large signal for the first gain stage that will be "bottle-necked" by the decoupling cap before the signal goes to the "Fuzz Face" part of the circuit.  Am I correct in this?  Is this also why a Tonebender Mk II uses a relatively large input cap (~ 4.7uF)?

If I just want to drive the Fuzz Face circuit a little harder -- not so much that it still can't clean up a bit when I roll back the guitar volume -- should I stick with the 10uF input cap as a reference point?  are there other options or strategies?

And one more thing...on the ZVex Fuzz Factory, the emitter from Q1 goes directly to ground -- am I right that if a resistor is placed between the emitter and ground, the gain is decreased (and with a bigger resistor, there's a proportiante decrease in gain)? 

[DavenPaget]

Haven't you seen GEOFEX ?

[jrod]

I would say 10uF would be a good reference point. The Marshall Supa Fuzz used a 10uF in that spot, too. But, I would think something in the 4u7 neighborhood would work well. It would be a good idea to breadboard the circuit and tweak that cap until you find exactly what you need. I've messed around with this type of thing before and got a lot of noise so that is something to consider. There are some silicon tone bender mkII circuits floating around that might give you some ideas. Check out Dragonfly's file in the gallery.

I don't about the zvex Q1 emitter resistor question, but if it does decrease the overall gain it might make a cool "pre-gain" type control.

[mordechai]

Haven't you seen GEOFEX ?

Well yes, but there's a TON of info on that site and it's difficult to find a specific answer easily. 

[mordechai]

I would say 10uF would be a good reference point. The Marshall Supa Fuzz used a 10uF in that spot, too. But, I would think something in the 4u7 neighborhood would work well. It would be a good idea to breadboard the circuit and tweak that cap until you find exactly what you need. I've messed around with this type of thing before and got a lot of noise so that is something to consider. There are some silicon tone bender mkII circuits floating around that might give you some ideas. Check out Dragonfly's file in the gallery.

I know, I am a little concerned with noise as well, but a 2n5087, which seems to be a prety quiet unit, might do the trick, as well as tweaking resistors later on in the circuit to gate-out some of the noise...

[Meanderthal]

 Using a very large input cap on a fuzz not only allows all the low end you can hear through, it also allows subsonics through as well. Which then get distorted, the harmonics of the subsonics are audible, so you get those weird splatty noise artifacts.

[iccaros]

Haven't you seen GEOFEX ?

Well yes, but there's a TON of info on that site and it's difficult to find a specific answer easily. 

correct
the input capacitor and the and the first resistor create a RC circuit, which rolls off frequencies. At 10uf you pass everything a guitar can send out, adjusting this would remove frequencies form the audio path, so if the design you has too much bass you can lower it to say a 2.2uf or a 1uf and see if you like it more.
the intended goal is to block any DC voltage coming in the pedal from an incorrectly created source, its a happy accident that it also can be used for tone shaping.  

[Gus]

Look at this for some ideas
http://www.aronnelson.com/gallery/main.php/v/DougH/fuzz/hotsi/GS_Tonebender.gif.html?g2_imageViewsIndex=1

First stage is about a gain of X10.  10K collector resistor into a FF type gain stage
note the Fat switch and the coupling cap values.

Look in schematics,  look at 3 transistor fuzzes, look in the galley(link at the top of this web page)  Search and study lots of schematics on the web.

[jrod]

Here is that si bender in the dragonfly folder.

It's got the 4u7 input and a 2k7 resistor from Q1 to ground.

[mordechai]

This is all very helpful, so thanks to those who have chimed in.  A few questions:

1. on the Gus Smalley "Hot Silicon", what is the purpose of the R4, 1K resistor to ground on the emitter of Q1?  What is the benefit as opposed to just tying the emitter to ground directly?

2. Same circuit, different question: what would removing R1 (150K), R2 (22K) or both do to the first gain stage?

3. on the Dragonfly "Silicon Tonebender", what would be the effect of removing the 100K to ground resistor right before the base of Q1?

4. Same circuit, different question: what purpose does the 100 Ohm series resistor between the power source and the 47uF cap serve?

5. Same circuit, different question: is the placing of the .022uF cap in parallel with the 1K along the power rail (between the emitter resistors of Q2 and Q3) for "smoothing" purposes because the transistor properties would otherwise yield to sharp a tone?

Again, I appreciate the help, folks.

[sault]

1. on the Gus Smalley "Hot Silicon", what is the purpose of the R4, 1K resistor to ground on the emitter of Q1?  What is the benefit as opposed to just tying the emitter to ground directly?

Top to bottom, the pinout on the 2n5089 facing away is C B E. The off-hand answer is that R4/R3 controls gain, R4/R3 = 10k/1k = 10x, and that's plenty enough. Any more than that and you're going to clip hard and just raise your noise floor unnecessarily.

R3 is your emitter resistor, and that helps stabilize against temperature. As the temperature raises, more current flows, and an emitter resistor means that consequently more voltage will drop, keeping your gain approximately stable.

2. Same circuit, different question: what would removing R1 (150K), R2 (22K) or both do to the first gain stage?
3. on the Dragonfly "Silicon Tonebender", what would be the effect of removing the 100K to ground resistor right before the base of Q1?

R1 || R2 provides the voltage/current to bias the transistor. If you remove them, your transistor basically doesn't work. Same with the Tonebender.

4. Same circuit, different question: what purpose does the 100 Ohm series resistor between the power source and the 47uF cap serve?

Filters noise/smooths the ripples from the power supply. I suppose its not 100% necessary with a battery, but in a high-gain circuit its good practice. I don't think it necessarily needs to be that big of a cap, but whatever.

5. Same circuit, different question: is the placing of the .022uF cap in parallel with the 1K along the power rail (between the emitter resistors of Q2 and Q3) for "smoothing" purposes because the transistor properties would otherwise yield to sharp a tone?

It's not obvious from the schematic, but the 1k and 5.6k resistors together make up Q3's collector resistance. Since the instructions call for biasing to 4.5v at the collector, splitting the resistance like that changes the output bias, meaning that the gain is kinda-sorta raised. Or at least that 2nd order harmonics are added. The cap acts a low-pass filter, yeah. 1k || 0.022uF = 7200 hz.

Hope that helps, a little.