AW: [sdiy] OTA's, Iabc, and the magic of V to I

[Magnus Danielson]

From: "Nils Pipenbrinck" <np at inverse-entertainment.de>
Subject: AW: [sdiy] OTA's, Iabc, and the magic of V to I
Date: Sun, 16 Dec 2001 19:52:40 +0100

> Hi Magnus (Scott, Scott and the lurkers),
> 
> > According to my datasheet shall the amplifier bias voltage vary from
> > about 500 mV (at 100 nA and +25 degree C) to about 700 mV (at 1 mA and
> > +25 degree C).
> 
> Wait wait wait..

No, I won't wait! ;O)

> Amplifier Bias Voltage?

Yeap!

> I thought Pin5 of the 3080 is a current input pin. All references I found
> are driving it with current. I know that I can also drive the bias input
> with a small voltage (which in turn will let a current flow), but isn't that
> the wrong way to do it?

You are correct, it IS a current input pin. But this does not exclude
that the voltage may change, does it?

My poing was really, as you change the current, beware that the
voltage of that pin indeed chages alongside with the change of the current.

> > You may not realize it, but the pin is allready connected to a current
> > mirror. In the schematic diagram of the chip D1 and Q3 forms a current
> > mirror. Maybe another current mirror provides sufficient isolation.
> 
> I didn't realized that. Stupid me thought I just provide some current into
> the base-emitter junction and the diode is just some kind of input
> protection. That changes things quite a bit and makes my current mirror
> approach pointless :)

The trick is that you don't perceive it as a current mirror due to the
way it is drawn, but if you know how a propper mirror looks and know
of common drawing-rules changes, it becomes clear that it indeed is a
current mirror. Besides the input diff-pair, the CA3080 is just full
of current mirrors, there are 4 of them in total!

> > Also note that the atanh curves from signal-input to output is due to
> > the diffrential pair (Q1 and Q2) and linearisation of that would
> > require additional diodes to some positive voltage and then feed with
> > current in order to logarithmize the input voltages.
> 
> Hm. I'm not sure about that. I didn't run the inputs hot. I just provided
> some few millivolts into the diff-amp input. (the maximum it saw was about
> 150mV, but the output was almost saturated at 50mV). Afaik the datasheet
> sais that anything below 300mV is fine for linear work.

Hot is a relative term. The gain and the "hottness" of the signal is
really the same thing and they are both controled by the common
emitter current. If you want me to I could toss you the equations for
the diff-pair and guide you through the rest of the CA3080...

So when you see a atanh curve that is according to theory of the
diff-pair, the current mirror is just doing it's job.

> But I am sure, that when I increased the input voltage, Iabc decreased, and
> the output saturated. I don't know if the decrease in Iabc is caused by the
> increasing output current or the input voltage, but I can find out by some
> tests - this time with a better current source (one that doesn't suck when
> the load changes).

Indeed.

> > To use the input diffamp current to vary the gain is a gross
> > oversimplification.
> 
> Not sure about that either. I don't have the schemantic of the 3080 here,
> but I remember the current mirror was just connected to the common emitter
> of the diff-amp.

Right, but my comment was more a generic one about the use of the
input diff pair for the variable gain stage may not be the best of
solutions.

> The current would directly translate into an amplified
> difference voltage across the collectors and thus control the gain (at least
> if I don't take the current flowing through the base-emitter junctions into
> account, but this current should be quite low if not neglible).

Right.

> > Just don't scratch your head too much on OTAs and current mirrors
> > then...
> 
> I'll have a lot of time to spend in hotel-rooms the next days.. scratching
> my head over current mirrors will be a lot of fun :)

I meant when doing the "buissness" part of the travel. I rarely get
much time of for private headscratching when doing my buissnesstravels
thougth. Usually too much good friendship and red french wine comes
inbetween ;O)

> Anyways - now I know, that a V to I converter just needs a tranny and an
> opamp, so I can go and verify my ideas on my testboard.
> 
> Thanks and Cheers,
> 
> 	Nils Pipenbrinck

Cheers,
Magnus