phase margin in closed loop feedback

Hi, i'm a little bit confused about phase margin in close loop

so let say i design an amplifier in an open loop has a phase margin of 70.

in close-loop, i break the loop and insert the feedback value that i want to inspect at the input, making the the circuit looking like a open loop again.
So the phase margin will go from 180 degree (because of negative feedback) to 0 degree. So the phase margin of a closed loop = 0 - (phase at gain = 0)?

is this correct?

thanks

I have some problems to understand the 2nd part of your question.
Nevertheless, perhaps the following is important for your understanding:
The phase margin concept is based on open-loop properties only.
That means: You simulate the magnitude as well as phase of the open loop (that is the so called "loop gain"). Then you determine the phase margin.
That`s all. This margin gives you a measure of stability for the closed-loop case.

In addition to what LvW has said, ultimately the location of "closed-loop" poles determine stability. But since it is difficult to estimate their locations and to create a merit for relative stability, "phase margin" is used as a tool that is computed from open-loop transfer function. Phase margin is interesting because it provides some information about "closed-loop" poles without much work. It is just that and there is no point calculating the phase margin for "closed-loop". What would be interesting to see on a closed loop is rather the peak overshoot & settling (time response) and resonant peak around UBG (freq response) if they exist that is.

saro_k_82 said:

........ Phase margin is interesting because it provides some information about "closed-loop" poles without much work. It is just that and there is no point calculating the phase margin for "closed-loop". .......

Click to expand...

Hi saro, for my opinion, the only information about closed-loop poles I can derive from the phase margin PM is as follows:

As small PM is an indication for the location of the dominant pole pair rather close to the Im axis of the s-plane (and vice versa).
Restriction: This holds only for loop gain functions that allow application of the PM concept - because there are other functions (depending on zero locations) that require the application of the complete Nyquist criterion.

Do you think we can derive more detailed information on the poles from the PM?

LvW said:

As small PM is an indication for the location of the dominant pole pair rather close to the Im axis of the s-plane (and vice versa).

Click to expand...

Agree on this LvW. Open-loop ac gain plot also provides some information about the y-component of the dominant pole pair (given that x is small), without having to work on the exact closed loop expression. My statement was based on this fact that a poor PM provides a nice estimate of both the x and y coordinates of the dominant pole pair correctly for most systems that we tend to fall upon.
I agree on the restriction too, but for systems other than the ones displaying multiple gain / phase crossovers, one can depend on just this information right.
BTW the question is about PM and not the way in which it is obtained. Nyquist provides PM which is easy to obtain as long as there are no open-loop right half plane poles, which I believe are far from use of analog IC designers.

Saro - full agreement.
LvW