Common Collector vs Common Emitter amplifier

[cdwillis]

Hi yall. Sorry if this is a lot of text. I've learned a ton reading through this forum and Jack Orman's Muzique.com website

I'm very new to building stompboxes. So far I've built the one knob Coloursound Fuzz on vero board, a Devi Ever Soda Meiser on vero board, and a Super Hard On clone on a PCB. I've been reading a lot trying to wrap my head around why the SHO works like it does and have been breadboarding LPB1s and Electras to experiment with biasing BJT transistors. I thought I was starting to understand how these things work as common collector amplifiers, then while trying to understand how to raise input impedance on these circuits while still biasing the transistors, I came across common emitter amplifier circuits. The article I was reading said that the CE circuit isn't that great at amplifying, but it is good at keeping a high input impedance. So a CE circuit is basically good for a buffer and a CC circuit is good for gain/amplification.

So my question, I suppose, is what's the point of the buffer in the front of an amplifier circuit? If the signal already has impedance loss from a cable it can't add that back, can it? I'm not sure if I'm misinterpreting what Jack said on his buffers page, but he said generally you want a high impedance input and a low impedance output. Is this right?

[PRR]

> The article I was reading said that the CE circuit isn't that great at amplifying, but it is good at keeping a high input impedance. So a CE circuit is basically good for a buffer and a CC circuit is good for gain/amplification.

I think you are mixed up. CE is by far the most common amplifier.

[cdwillis]

Yeah, I'm probably all turned around at this point and mixed up which circuit was which. I guess what I'm trying to ask is that I don't understand why I would want the high input impedance, but low on the output?

[imJonWain]

It depends on the situation.  Generally speaking (with pedals) it's so the circuits interact with each other as little as little as possible.  The explanation depends on how in depth you want to get.

https://electronics.stackexchange.com/questions/21787/why-is-high-input-impedance-good

[duck_arse]

the low impedance output will drive stuff, like loads. the high impedance inputs present small loads, less trouble for outputs to drive.


I'm just replying cause I wanna say it's neat that your suit has markings for reflection reading.

[PRR]

> I don't understand why I would want the high input impedance, but low on the output?

I get power from (somewhere, utility, battery, squirrel-wheel?). I have lights and appliances. I want to plug stuff in and in random combinations without lights dimming.

If the source is much lower impedance than any or all of the available loads, the lights will not dim.

ZERO source and INFINITE load are often over-kill, even impossible.

My power wire from the street is about 0.4 Ohms (sags 8V for 20A load).

Typical loads, ratio to source impedance, and delivered voltage:
12W LED light: 1,200 Ohms, 3000:1, 125V
120 Watt under-desk heater: 120 Ohms, 300:1, 124.5V
1,200 Watt room heater: 12 Ohms, 30:1, 121V
Well-pump 40A starting surge: 3 Ohms, 7.5:1, 110V

Turning-on the lights or the desk heater causes "no" lamp dim. The room heater makes a very slight dim but you have to be watching for it. I can't even hear the underground well pump start but the lamp-dim is noticeable anywhere in the house.

[cdwillis]

Thanks. I think I'm starting to get it. I'm building a son of screamer right now and wonder what kind of effect not having the buffers from the stock screamer circuit is going to do.

[R.G.]

Here's another view of the same information.
First, you have to consider both input and output impedances. A source impedance of zero effectively says "I don't care how much current you need, I can force the voltage to be what I want." Real sources are never that capable.

Electronically, real signal sources, AC or DC, act like there is a "perfect source" hidden behind an internal series resistor, series inductor, and shunt capacitor, but where you cannot get at them. Sometimes this is a real part of the signal source, sometimes not. For instance, a single coil pickup acts like a signal voltage of some size, but hidden behind the real resistance of the windings (commonly about 8K) and the inductance of the windings, often 1H-2H.

The big kicker in guitar pickups is usually that inductance. An inductor has (ideally) zero resistance at DC, but its impedance increases linearly with frequency. A 1H inductor has an impedance of Xl=2*pi*L*f, where f is the frequency. So at 1kHz the impedance of a 1H (common for a pickup, remember) is 2*3.1416*1*1000= 6.283k. At 10kHz, it's 62.83k.

A source isn't any good all by itself, so it ought to drive something. This is where input impedance comes in. Every amplifier input acts like it has an impedance to ground inside it. This is just a reflection that if you force some voltage onto an amplifier input, it "eats" some current from the signal to make it amplify. To make this worse, any biasing resistors on the input to the amplifier eat some signal current too. This winds up being conceptualized as the input impedance of the amplifier: literally, the ratio of signal voltage at the input to the current going into the amplifier, expressed as R = V/I.

Neither input nor output impedance means anything without the other, and it's how they work together that we manipulate. We mostly want to transfer all of the signal voltage from source to amplifier that we can; we don't want to lose any. The output impedance of a source, a resistor or inductance, feeds an input resistance to ground. A little math and Ohm's Law tells us that this acts like a voltage divider, and cuts down the input voltage that the amplifier actually sees. The ratio of the impedances tells us how much we lose. Little loss is always better if you can work it, so we want the lowest output impedance sources and the highest input impedance loads that we can get if we're looking for best signal voltage transfer.

A rule of thumb is that for best voltage transfer, you want the source output impedance to be less than 1/10 of the input impedance/load it feeds. 1/100th or 1/1000th is better, but try to get at least 1/10th.

Guitar pickups are a special case because of that great honking inductance. It's not so much that inductance is bad, but the changing impedance of the inductance with frequency means that the inductor impedance goes up with frequency, so a guitar pickup will have a variable signal loss driving a load. The highs are proportionately more divided down than the lows. This makes for a duller sounding signal, and is the origin of the term "tone sucking" as relates to guitar gear.

Amplifiers are sources too - they all have outputs. We tend to want amplifiers to have high input impedance so they can "accept" any signal source without loading it down, and also to have a low output impedance so they can drive anything we connect to them. The common collector, also known as the emitter follower, does this better than any of the three possible single transistor arrangements. It's only failing is its voltage gain - always slightly less than one. So you lose a little bit of your voltage, but you get good input impedance and good output impedance.

The common emitter has a lower input impedance and higher output impedance. It's advantage is that it gives voltage gain.

It is very common to use a CC/follower on an input, especially for guitar circuits which have to cope with that tone-sucking inductance, and have that then drive a CE or other amplifier which might have a lower input impedance, but provide the necessary voltage gain.

This discussion is all about voltage amplifiers and sources; for current input/output amplifiers, you want exactly the reverse. But we don't use current-mode amplifiers much.

[antonis]

Another whipstsitch for R.G. been neither out of time nor out of mood.. 

[cdwillis]

Thanks, RG. You have no idea how much I'm learning by reading your stuff. And thanks everybody else too. I have a long way to go, but it is starting to make more sense to me.