[sdiy] Simple discrete Unity-Gain Follower ?

[Magnus Danielson]

From: Michael Buchstaller <buchi at takeonetech.de>
Subject: [sdiy] Simple discrete Unity-Gain Follower ?
Date: Wed, 30 Apr 2003 17:11:07 +0200

> Hello Friends,
> 
> i am looking for a simple circuit t decrease the output impedance
> of my synth modules.
> OK, i can just use an Opamp with the "-" input tied to the output to buffer
> the signal, but i would like to make a discrete transistor output stage.
> An Emitter follower won´t do because of it´s Offset and unequal push/pull
> currents.
> 
> Any ideas or pointers to a simple solution ?
> 
> It needs not to deliver lots of current, but should be capable of driving several
> 600 Ohm inputs in parallel. (say: 150 Ohms input impedance)
> 
> Could it be someting simple like:
> 
>                        +15V
>                          ^
>                          |
>                          / E
>                        |/
>                  |-----|
>                  |     |\
>                  |       \ C
>                  |       |
>                  |       |
>  Input ()--------|       |-------------()  Output
>                  |       |
>                  |       |
>                  |       / E
>                  |     |/
>                  |-----|
>                        |\
>                          \ C
>                          |
>                          |
>                          V
>                         -15 V

That one will not really cut it for linear purposes, since there will a gray
zone where the input is between the points where either of the upper and lower
transistors is really biased. This causes really low current/high impedance as
well and you can't really say you have a good drive there. The remedy is to
insert a bias network that will impose a bias voltage onto the input signal.
There are many ways to acheive this. The bias-voltage can be trimmed not only
to minimize the area where the neither the transistors is properly biased and
conducting, but you can get overlapping areas. This way you have moved from a
class C stage, over to Class B and into class AB. If you make both transistors
conduct over the full range, then you have a class A. (Sorry if I have mixed my
class definitions up, but this is how I recall it at least, I don't think I'm
off by much anyway).

Then naturally, you can always toss a feedback-loop onto it, but that brings
a whole new realm of problems.

So, I guess this is why op-amps got so popular, they save you a bundle of
headaches. However, a properly designed amplifier may very well be op-amp free,
no problems!

The emitter-follower isn't that bad either, it have done marvels in the
computer field (CDC 6600, 7600, 8600 and then the Cray-1 to mention a few).

Cheers,
Magnus