Gain knobs in transistors

[midwayfair]

I'm trying to learn a little more about exactly how transistors do their thing.

I've observed that there are three ways to vary the gain and volume of a bipolar transistor. I'd like some tips on when using each is appropriate and most effective.

I don't really expect some of my questions to be fully answered, but if you can link me to some reading material for anything specific, that would be great.

1. "Pre gain" resistor between the input and the base after the AC coupling cap.
    This is what's used in the Lavache and the Fuzz Face. This is the one I'm most familiar with. It reduces signal going into the transistor, which can keep the input signal from overdriving the transistor. remove the resistance and you have more total signal. Is there anything else going on here?

2. Connecting the emiter to ground.
    This seems to be the most common type of control in transistor circuits, like the Fuzz Face, SHO and any number of boosts, etc. A pot or resistor is placed in series with the emitter and ground. Reducing the resistance increases the gain (and volume). It also causes the transistor to saturate.
    Two things I'd like to understand: (a) What's the large capacitor doing that's usually in these gain sections? I know it makes things sound different when you change its value, and I suspect it has something to do with protecting the transistor from shorts. (b) Is this used specifically because it produces the most amount of clean amplification?

3. Sagging the voltage source
    This is a series resistor or pot between the 9v source and the collector.
    This is one that really mystifies me. I know what sagging the voltage does (and I usually like it), but sometimes it's actually used as a gain control, like in Beavis's Trotsky Drive. When it's used there, turning it up (and removing resistance) not only gives more volume -- which makes sense to me -- but also seems to result in more distortion - which really doesn't make sense to me. I would think that lower voltage would mean that less resistance is needed on the emitter to cause the transistor to saturate.
     I've also noticed that sagging the battery sometimes results in some extra compression. Why is that?

[TNblueshawk]

Bump, good questions.

I've been wondering what the "dying battery" is all about. What does it do that changes the sound to where people actually like the sound better? I guess this is the same thing as sag? Not?

[midwayfair]

Bump, good questions.

I've been wondering what the "dying battery" is all about. What does it do that changes the sound to where people actually like the sound better? I guess this is the same thing as sag? Not?

It's one of my favorite effects, even though I don't really understand how it does all of the following stuff. Reducing the voltage does a few things to my ears:

1) Darker tone. Sagging the battery some is good for circuits where the treble would rip your ears off otherwise.
2) More compression. I expect this comes from being able to saturate the transistor easier.
3) More distortion without more volume, especially when you're stacking the pedal.
4) Lower total volume or lower headroom. There are some circuits where you want to lower the volume, but there are other places where it's less than ideal. For instance, if you don't want to load the input or output or reduce the gain, there's really only one place left to decrease the overall available volume in a transistor-based circuit: the power source.
5) In some situations, you can cause gating, spitting, and even stuttering, like by using it to misbias a fuzz face. (The bias trim on a fuzz face is controlling the voltage on one transistor, so it's doing the dying battery simulation on just part of the circuit). This is cool if you want the 60's garage rock sound. Of course, it can be done where it sounds horrible, too ...

[calciferspit]

I've played with starve on almost every circuit i've built, (to varying degrees of "keep it" or "lose it"), and I feel the best way to describe what's going on is a reduction of total wave frequency in the circuit. Whatever that I just made up means. Especially in fuzzes, the tone gets darker and the waveforms fatter. Reaction time slows, making it easier to pass lower freqs. Once it gets below the trigger of the guitar, it tends to oscillate and "play itself" decreasing in audio frequency as you starve it more. I've found the most use for starve in already crazy fuzz circuits, (blue box, devi evers, uglyface). not answering your questions at ALL, just a starve fetishist.