[sdiy] discrete SSM2018?

[Czech Martin]

Oh well, now the SSM2018 chip seems to be short in supply, or no
longer available. Perhaps we can learn the principles of operation
and create something similar, if not better.

Hello feedback haters, here's an idea to flame:

we can learn from said chip how to arrange feedback arround a gain core.
Usually you have feedback arround more or less fixed gain stages,
with all the good effects: less distortion, accurate gain, less
phase shift, output offset equals input offset voltage (for unity gain).

Now , using a variable gain stage inside a feedback loop obviously
creates some problem: if the gain is high all is well, but as
the gain decreases the now defunct feedback loop will no longer prevent
the input pair from clipping and so on.

Therefore the idea is to use two gain cores and to feed them
with dependend tail currents: a*I and (1-a)*I {a: 0...1}
so that the sum is always constant I.
If we now add up the signals of both gain cores, this sum signal
will also be constant. Now we are able to create a feedback loop.

This could be translated into using two of the 13700 otas in parallel
with a current splitter pnp differential amp as feeder,
or using two convetional gain cores or even a two quadrant multiplier,
as it contains two gain cores and a current splitter.

The later arrangement is the idea of the 2018 chip. See data sheet.

There will certainly be some stability problems, but the general idea should
work. The hope is that lower control feedthrough and output offset shift
can be obtained as well as less distortion and noise.
Also the operation can be designed to be linear as well as pseudo
exponential.
I would prefer linear since then only an expo converter has to be added,
for having both modes,
rather then a log circuit for the opposite case, which was a bit
overcomplicated as you can see in one of my schematics.


comments?

m.c.