[sdiy] Using both OTAs in 13700 as one OTA

[ChristianH]

Um, I didn't follow Aarons description in depth, but on cursory reading
I thought that's what he's trying to achieve - cancelling the offsets
using + and - inputs, while using the same Iabc on both.

Chris



On Thu, 4 Feb 2016 00:57:57 -0500 Michael E Caloroso
<mec.forumreader at gmail.com> wrote:

> Watch the offsets at the input pins.  They vary with iBias and can't
> be trimmed out.
> 
> OTAs are noisy.  Paralleling them would make a noisier system, it
> would not be differential.
> 
> MC
> 
> On 2/4/16, Magnus Danielson <magnus at rubidium.dyndns.org> wrote:
> > Aaron,
> >
> > On 02/04/2016 04:28 AM, Lanterman, Aaron wrote:
> >> Suppose you only needed one OTA to be a VCA, and you have a 13700 handy,
> >> which has two OTAs in it.
> >>
> >> Instead of using just one OTA, and leaving the other idle, what about a
> >> scheme like this:
> >>
> >> Let’s call the OTAs Negator and Positoria.
> >>
> >> Take the voltage input Vin and run it through the usual divide-down
> >> resistor divider, take that voltage and attach it to the negative input of
> >> OTA Negator,  and also attach it to the positive input of OTA Positoria.
> >> Tie the other inputs to ground.
> >>
> >> Take the current output of OTA Positoria, run it through a resistor of R
> >> ohms to ground to generate a voltage, call it V+, and attach that voltage
> >> to the positive input of an op amp.
> >>
> >> Take the current output of OTA Negator, attach it to the negative input of
> >> an op amp, and run a resistor of R ohms from the op-amp output to the
> >> negative input.
> >>
> >> The op-amp output is
> >>
> >> Vout = 2 R gm Vin,
> >>
> >> Where gm is the transconductance gain of the OTAs, assuming they’re the
> >> same.
> >
> > I would wire up a normal diffrential op-amp setup there.
> >
> >> I was thinking this might help cancel out the control signal bleed
> >> through, and might also have some SNR advantages. Or, there could be
> >> disadvantages that would counteract those possible advantages.
> >
> > Bleedthrough? Maybe to some degree for the common mode inbalance, but
> > lack of nulling biases doesn't help. Trying to create a better OTA bias
> > situation would help more, such as both inputs seeing the same impedance
> > and then a trimmer for offset compensation helps.
> >
> > SNR, sure, you doubled the transistor area for input. This gives you 3
> > dB improvement.
> >
> >> Anyway… does this sound like a reasonable thing to do, or a not so good
> >> thing to do?
> >
> > I would trim the sails somewhat differently if I would do it. I would
> > rather figure out what a second OTA could do for me in the design.
> >
> > Cheers,
> > Magnus
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at dropmix.xs4all.nl
> > http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> >
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy