Questions about hybrid Fuzz Face biasing

[DS13]

According to Electrosmash among other sources Q1c / Q2b should be ~0.7v with a germanium pair in a "perfect" Fuzz Face. With a silicon FF as I understand, Q1c would typically be ~1.2v - 1.4v.

But what about a hybrid fuzz face with low gain silicon Q1 and low gain Q2 germanium (Q2c is at 4.5v)? What is the "perfect" compromise / sweet spot of Q1c / Q2b voltage? In my trials, I am finding with limited si / ge combos with a 33K bias resistor coming in around 1.0v - 1.7v, and can adjust the higher ones down with a 24K.

I think I am finding the best performance with Q1c down a bit, around 1.1v with a 33K resistor.

For example a 2N4123 lands right in that ~1.1v range with a 33K, a 2N2369 as a few tenths higher, but gets closer with a 24K.

Realizing from a purely electronics POV, this is simply Ohms law, but from an audio performance POV, are my ears hearing correctly to bias Q1c slightly lower in this hybrid fuzz, than with a silicon pair? How do we determine the common Q1c / Q2b voltage for both to operate at peak performance?

Secondly, other than having less headroom, what is happening with Q1 when we bias it in the higher end of the range?

Thanks in advance..

 

[giselefortner]

But what about a hybrid fuzz face with low gain silicon Q1 and low gain Q2 germanium (Q2c is at 4.5v)? What is the "perfect" compromise / sweet spot of Q1c / Q2b voltage? In my trials, I am finding with limited si / ge combos with a 33K bias resistor coming in around 1.0v - 1.7v, and can adjust the higher ones down with a 24K.

I think I am finding the best performance with Q1c down a bit, around 1.1v with a 33K resistor.

For example a 2N4123 lands right in that ~1.1v range with a 33K, a 2N2369 as a few tenths higher, but gets closer with a 24K.

Realizing from a purely electronics POV, this is simply Ohms law, but from an audio performance POV, are my ears hearing correctly to bias Q1c slightly lower in this hybrid fuzz, than with a silicon pair? How do we determine the common Q1c / Q2b voltage for both to operate at peak performance?

Secondly, other than having less headroom, what is happening with Q1 when we bias it in the higher end of the range?

Thanks in advance.. geometry dash scratch

Hello,
There’s no single perfect number because it depends on the gain (hFE) of each device and how much headroom Q2 needs. But most builders bias Q2c to 4.5V first (as usual) and then tweak Q1c/Q2b by ear.

Empirical sweet spot:

• Hybrid FFs tend to sound best with Q1c / Q2b ~1.0–1.2V
• High enough for Si Q1 to swing cleanly
• Low enough to keep Ge Q2 happy (and not starved)

Your observation that ~1.1V with a 33K bias resistor sounds good is right on target. If you raise it up near 1.4–1.5V (typical silicon value), Ge Q2 starts to sound brittle/thin because it’s essentially “over-biased” and loses the soft gating/sag qualities.

Moreover, when Q1c creeps up to ~1.5–1.7V:

Less headroom for Q1: Its collector is closer to the supply, so the input signal clips earlier and more asymmetrically. This usually sounds sharper and harsher.

Pushes more current into Q2:
- Increases bias current through Q2 base
- Can overdrive Q2, causing it to saturate/clamp
- The fuzz gets compressed and can lose the dynamic "bloom"
- Noise floor rises: More current = more hiss

On the flip side, if you bias Q1 too low (<1V in this hybrid), you get:

- Starved Q2 (gated decay, sputtery, volume drop)
- Loss of bass response because the transistor isn’t in its linear-ish range