Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

why neg. feedback converts into positive feedback ?

Status
Not open for further replies.

desperado1

Full Member level 3
Full Member level 3
Joined
Sep 4, 2006
Messages
189
Helped
8
Reputation
16
Reaction score
2
Trophy points
1,298
Activity points
2,503
i have read that every negative feedback converts into positive feedback at higher frequencies. How is that possible ?? i m totally confused........
plz help me
 

At higher frequencies phase shift can become 180 degrees and negative feedback will turn positive. Therefore opams require phase margin. It's mean when gain of operational amplifier lower to 0 dB phase shift should be (180 deg. - "phase margin").

So, not every negative feedback converts to positive, only poor. =)
 

can u please elaborate on this point
It's mean when gain of operational amplifier lower to 0 dB phase shift should be (180 deg. - "phase margin").
 

Gain of opamp decrease when frequency increase. And phase shift increase when frequency increase.
When gain decreace to value 1 (0 dB) you should have a phase margin, otherwise when you use your opamp with negative feedback it can turns in positive.

Phase margin define as 180 degreess minus phase shift at unity gain frequency. Unity gain frequency is the frequency where gain = 0 dB.
 

Warlike said:
...........................
So, not every negative feedback converts to positive, only poor. =)

In real systems there are always phase shifts of -180 deg. above a certain frequency.
Thus, every feedback which is negative at lower frequencies will turn into pos. feedback above a certain frequency limit. However, that´s not catastrophic if the magnitude is small enough.
This effect was described by Bode already more than 50 years ago in connection with the so called "water-bed-effect".
 

LvW, I mean that in good design conversion of negative feedback to positive feedback is not critical, like in poor design. And one can say that neg. feedback don't turns into pos. if it will not cause problems.

But if we will operate in strictly terms - yes, every neg. feedback turns into pos.
 

As I understand it......there are various resistor, capacitor networks that are formed in a circuit both from actual circuit elements and parasitics. These can give you two types of RC networks, those that give single pole response and those that are in the feedforward path and give a right half plane zero type response. The single zero responses are the culprits that cause your phase to plunge towards
-180 at higher frequencies. When phase tends towards -180 you get an unstable system because it means that a negative feedback is changing to positive feedback. In fact that is why we might need to add some miller compensation, which is adding a RC in a some feedforward path, like from a gate to a drain to improve the phase at higher frequncies thus giving you phase margin to improve stability of the system.

This is my simplistic explanation. Please feel free to comment or criticize.
 

As I understand it......there are various resistor, capacitor networks that are formed in a circuit both from actual circuit elements and parasitics.
Right !
These can give you two types of RC networks, those that give single pole response and those that are in the feedforward path and give a right half plane zero type response.

For real systems there will be no single pole response because of parasitic effects.

The single zero responses are the culprits that cause your phase to plunge towards -180 at higher frequencies. When phase tends towards -180 you get an unstable system because it means that a negative feedback is changing to positive feedback.

No, this effect will lead to instability only if the magnitude of the loop gain is > 0 dB at this 180-deg-frequency.

In fact that is why we might need to add some miller compensation, which is adding a RC in a some feedforward path, like from a gate to a drain to improve the phase at higher frequncies thus giving you phase margin to improve stability of the system.

No, it´s the wrong explanation. The Miller effect causes an additional pole which is dominant and shifts the phase at lower frequencies (principle of single pole response, but at the cost of a smaller bandwidth).
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top