# OpAmp Non-standard Feedback

1. Feb 22, 2013

### Rudinhoob

OpAmp "Non-standard" Feedback

Hello,

This is my first post here. Please excuse my poor English and grammar. I will try to make my questions clear.

As I understand, with a negative feedback, where the output connected to the inverting input and in the ideal case, the Vn and Vp will be very close in a sense that Vn tracks the voltage of Vp so that the difference is almost zero.
However, when changing the feedback configuration, for example connecting the Vn input to other parts than the output, or using a positive feedback, how does this affect the ideal rules in analysis?

Thanks.

2. Feb 22, 2013

### jim hardy

Re: OpAmp "Non-standard" Feedback

That's a hard concept to wrap your brain around when first starting out.

The little integrated circuit we call an "op-amp" is just a high gain amplifier.

When we surround an "op-amp" with a feedback circuit that allows it to hold its inputs equal,
we now have an electronic circuit that behaves in accordance with some mathematical "operation" defined by the feedback network.
So we call it an 'operational amplifier' circuit.
Perhaps we could have done better with the naming comvention, perhaps not.

The circuit you describe , signal applied to +in and Vout connected to -in, performs the math operation Vout = Vin - 0.
More precisely, Vout = Vin(1-1/(amplifier's open loop gain)),
and since open loop gain is around a million that reduces to ~Vin - Vin/(106), whhich rounds to Vin - 0.

Remember - it is the duty of the circuit designer to surround his amplifier with a feedback circuit that allows it to hold its inputs equal.
Else it won't operate.

See if this is any help in your studies:
http://www.analog.com/static/import... Amp Applications Book (PDF)/P2 ChH_final.pdf

This site is a treasure trove:
http://www.philbrickarchive.org/

old jim

Last edited: Feb 22, 2013
3. Feb 25, 2013

### Rudinhoob

Re: OpAmp "Non-standard" Feedback

How can we tell that a certain feedback configuration makes the two input voltages almost equal, and hence assume Vn = Vp?

4. Feb 25, 2013

### jim hardy

Re: OpAmp "Non-standard" Feedback

What i do is write Kirchoff's voltage law solutions for voltage at each input, then set those equal and solve for Vout.
If Vout doesn't fall between the supply rails, the circuit won't "operate".