"siamese twin servo"

[jeffrey_burr]

So the idea I'm toying with is a Twin Servo mic amp but with the "shared gain" configuration brought to my attention courtesy of Samuel Groner.  I guess I'm pretty tickled by that design.  The first stage could be servo-ed just like the Jensen app note shows.  For the second stage, um, an inverting servo directly feeding the non-inverting input ("doubly virtual earth"?) of the gain stage, or, uh.... how does that work, servo-ing an inverting opamp stage?  I guess the easy answer is, use modern jfet opamps and fuggedaboudit, but I have to use these discrete bipolar numbers for something.  Sorry if this is way beyond stupid.

 

[Samuel Groner]

http://www.groupdiy.com/index.php?topic=38316.0

Samuel

 

[jeffrey_burr]

....and there he is.  Good day and thanks for the link Mr. Groner.  My idea was a little bit different, based more directly from the Jensen application note and using component values from the shared gain schematic you share on your own site.  In other words, not "my idea" at all!  So the servos would be trimming the non-inverting inputs in both halves, but I'm not certain exactly how that works with the inverting stage. You've worked it out for opa627's and I guess 990s or similar devices would have different noise and loading issues.  I'll keep banging my head against it for a while, and maybe make a drawing. 

 

[Samuel Groner]

I'm not certain exactly how that works with the inverting stage.

It doesn't matter if its an inverting or noninverting stage (other than for the inverting stage you want to add a voltage divider at the noninverting input in order to decouple servo noise); what matters is if you want to feed the servo output to the inverting or noninverting input. The former requires a noninverting servo, the later an inverting servo.

But make sure to read the other thread; it ain't easy to keep any DC from the pot.

Samuel

 

[jeffrey_burr]

from that thread:

The only way to get all DC out of the gain pot is to apply the servo to trim up/down the + input so opamp - and output is sitting at 0V.

Of course this means you will be putting DC at the transformer assuming all that is DC coupled.

Well that doesn't sound so hard, in fact this is the way I had pictured doing it, as it appears on the Jensen application note: an inverting servo trims the non-inverting input, through the secondary of the input transformer (feeding the "earth" side of the winding if that makes any sense).  My brain won't yet tell me if this solves anything though.  Patience.