[sdiy] Operational Transconductance Amplifiers

[Magnus Danielson]

Walker Shurlds wrote:
> Recent post called Source for 3080's reminded me, so I thought I'd ask.
> 
> As a noob, albeit one with circuit analysis skills, I've decided to
> analyze every part of the ASM-1 until I grok it before building
> anything serious.

This is a good approach if you got the skills. There are a number of 
challenges. The VCO and VCF having most of them.

> There are a few minor things I don't quite get yet
> (the capacitor between the negative input and the output of the expo
> converter op-amp, for instance) but I'll figure it out in time.

Limiting the bandwidth for the sake of stability.

> But when I flipped the page from the VCO to the VCF I said "WHAT THE
> HELL IS THAT!?".  Eventually figured out that it's one of these OTA
> things.
> 
> Anyway, I'm having difficulty figuring out how they work.  Wikipedia
> gave me an equation for the open-loop gain (Iout = Vdiff*gm), but that
> doesn't tell me at all what Iabc or Ibias are for at all, and doesn't
> help much with solving a circuit. Based on the ASM-1 VCF schematic, Iabc
> is the part that actually does the "VC" in "VCF"!

The Iout = Vdiff*gm part is the magical transconductance part, but OTA 
has a twist to it... the gm part can be varied by a current, the Iabc 
current.

Thus, the effective formula becomes Iout = Vdiff*Iabc*const. The input 
voltage will flipp the direction of the output current but the Iabc 
current does not allow it, it only scales the output linearly (well... 
nothing is really linear but as the basic concept).

> So, yeah.  How exactly do these things work and/or does anyone know of
> a good website/book that explains them thoroughly?

Not too complex. Behaves like a current-controlled resistor when looking 
into a virtual ground, which is what they do in the VCF.

Cheers,
Magnus