Fuzz face caps question

[jimmyandrex]

Recently I posted a question about Ge hfe/leakage in a fuzz face circuit and recieved some very useful forum advice about biasing the resistors;

Here are the results for Q1 50hfe   Q2 100 hfe ( taking into account leakage reading) using trim pots

a)Q1 14.4K -0.5v *
  Q2   9.59K  -4.8v

b)Q1  12K   -0.8v
   Q2   3.5K -6.5v

But at:

http://www.diystompboxes.com/analogalchemy/emh/emh.html

using the fuzz face calculator I got completley different results for the result marked*
which was 47.5K for Q1 at -0.5v.

I rotated both trim pots either way to cutoff to test the results aurally
Those readings did sound best with the higher (b) being the best (JBF-3 voltage reading)
So I'm sure they are right by listening tests.
But it did sound quite flubby and fuzzy with the neck pick up and guitar vol pot on full though
I'm considering reducing the value of the capacitor on either the input 2.2uf coupling cap (or should it be the 22uf cap on the fuzz control) to reduce the bass
So there are few questions here - any further advice would be appreciated.

[kaycee]

You can try reducing the input capacitor, say 100n. You can also play with the output capacitor and the size of the volume pot to change the tone. FFs are a bit wooly which is why some people can't get along with them, personally they are still my favourite fuzz.

It's worth experimenting with the in and output capacitors, you get different sounds when the in cap is bigger than the out and vice versa, going down to surprisingly small values. The vol pot can vary from 50k to 500k which also has an effect on the tone. Can't remember which way cuts bass though!

[jimmyandrex]

100k is recommended for the vol pot I have found- but is there a more significant adjustment elsewhere which could tightened the bass issue a tad but retain the tone characteristics/dynamics? 

[Davelectro]

In my experience, 47nf is the perfect choice for the input capacitor if using humbuckers.

BTW, I've never found the EMH calculator particularly accurate. Not even with Si transistors.

[Electric Warrior]

Any reason you picked such a high hfe transistor for Q2? In my experience you need to match the low gainers rather closely to make them sound good with stock values.

[polifemo]

- Lowering the input cap will make the pedal "tighter sounding" but if one goes too far the fuzz character will go away and the fuzz box will sound more like a distortion pedal.

- Lowering the output cap will keep the character of the fuzz but with a controlled low end.

-Lowering the fuzz cap is a way to change the character of the pedal, but if going to small the clean up when lowering the guitars volume will be too drastic. In a FF, with more or less standard values I feel that 10uF is as small as one can go and that 15uF is better.
I usually keep the 22uF though and feel that 10uF-15uF works best in high gain FF's.

You could try: 1uF input cap, 0.01uF output cap and 100k volume pot as a starting point

[jimmyandrex]

Electric Warrior - Geofex article recommends "If you allow combinations of one high and one low gain device, the range widens out to 70 or so on the low end and perhaps 130 on the high end" so I chose the ones nearest in ratio to that , I the arbiter JFB-3 has low and high combination -I think thats the best sound I have heard from an FF.. I'll certainly try matching the low hfe ones as an experiment.
Polifemo/Davelectro-
I'll experiment with matched hfe first and maybe
I'll try lowering the output cap later

[Electric Warrior]

With Q1 as high as 70 or 80 you can get away with a wider spread in my experience. But it's a matter of preferance to some degree.

When you match them closer, you could also try transistors with different amounts of leakage, if you have a couple to choose from.
Just use your ears 

[Davelectro]

- Lowering the input cap will make the pedal "tighter sounding" but if one goes too far the fuzz character will go away and the fuzz box will sound more like a distortion pedal.

This is true. With 47nF you start to get less of a fuzz and more of an overdrive, so to speak. A nice heavy overdrive.

[MetalGod]

Add some series resistance to the input (as if you were turning your guitar pot down) and that tightens things up (and cleans it up a little bit too).

I prefer 0.1uF - 22nF for the FF input cap, but I use humbuckers.

[pinkjimiphoton]

With Q1 as high as 70 or 80 you can get away with a wider spread in my experience. But it's a matter of preferance to some degree.

When you match them closer, you could also try transistors with different amounts of leakage, if you have a couple to choose from.
Just use your ears 

+1

[jimmyandrex]

Ive experimented with two with similar gain ac128 with similar acceptable leakage (50hfe) however the subjective aural test followed by meter readings bore no relationship to the "usual" recommended voltages (which didn't sound at all good)

Q1  8.1K  bias resistance    -1.61

Q2  2.0K  bias resistance     -5.59

At these settings the sound test was quite acceptable ranging from clean(ish) to massive OD which didn't sound too woolly fuzzy/farty . I think I can improve upon this with adjusting the input cap etc as advised.

The voltages and resistance readings appear quite wrong though

[Electric Warrior]

Yeah, the resistors are quite a bit off. Voltages are alright.
How much do your transistors leak?

[polifemo]

Ive experimented with two with similar gain ac128 with similar acceptable leakage (50hfe) however the subjective aural test followed by meter readings bore no relationship to the "usual" recommended voltages (which didn't sound at all good)

Q1  8.1K  bias resistance    -1.61

Q2  2.0K  bias resistance     -5.59

At these settings the sound test was quite acceptable ranging from clean(ish) to massive OD which didn't sound too woolly fuzzy/farty . I think I can improve upon this with adjusting the input cap etc as advised.

The voltages and resistance readings appear quite wrong though

I don't get your "numbers"...

The resistor on Q1's collector* should be somewhere around 33k (I would start there) and with a feedback resistor - from Q1's base to Q2's emitter - of about 100k (I would start there) you can then use whatever needed to get 4.5v (or whatever you might prefer) on Q2's collector.

* Q1 usually sits at about -0.5v

What's the leakage of Q1?

[jimmyandrex]

I am using bias pots for each transistor on these tests, however I have to admit that I didn't remove the transistors before taking the readings which I realise now has shown a false result because of leakage.

(For the test below I swopped around the transistor positions from the previous test.)

Q1 198ua leakage 51hfe
Q2 100ua leakage 48hfe

Using bias pots to adjust to recommended voltage

Q1   33k       -0.5
Q2   11.5k     -4.5

Aural test using bias pot adjustment for "best sound to my ears"

Q1    6.3k      -2.12
Q2    936R    -6.56

Q1 and Q2 removed from sockets before pot resistance reading!

[Electric Warrior]

Q1 198ua leakage 51hfe
Q2 100ua leakage 48hfe

Using bias pots to adjust to recommended voltage

Q1   33k       -0.5
Q2   11.5k     -4.5

Your transistors seem to be doing just fine in a stock fuzz face setup
Where does Q2C end up when you bring the resistor down to 8k2? And more importantly: what does it sound like?

[jimmyandrex]

And more importantly: what does it sound like?

Difficult to describe!
The stock values with these transistors sounds like the kind of fuzz sound I hate. I've built them before and they have all sounded crap whether silicon or ge - a sort of 70's amateurish cheesy fuzzy wuzzy tone. The sound I have arrived through the "different" or wrong setting is getting bit "closer" to what I would rather hear.

[jimmyandrex]

I think I will regard this as a partial success at the moment and put it on the back burner until I source some more transistors to test