Strange sharp rise/fall in stb phase plot

Dear SpiceMonkey,

SpiceMonkey said:

Sometimes phase rises sharply and sometimes falls sharply, while their magnitude basicaly has no difference.Changing sweep points density, probe location or coner...all these may lead to this "controdictory" result.

Could anyone help explan this phonomeon? thank you!

Let me know if you think I am misreading your plot of magnitude and phase for the two cases you illustrate, but it certainly appears to me that there is a difference in magnitude that may lead to the two corresponding difference loop phase results. Please recall, first of all, that phase changes are far more sensitive to the frequency than magnitude changes. Hence, it is far more likely you find more significant phase changes than magnitude changes as frequency increases. Secondly, examining your magnitude plot around 10 MHz, there appears to be a significant resonance in the blue curve that is not as evident in the red curve. As a result, I might expect a much larger phase shift associated with the blue curve than with the red curve - and that is what you are observing due to the phaseWrapped() function used by the Calculator. Hence, in my opinion - and perhaps ny thoughts are incorrect in your opinion, your phase results are not totally unexpected to me and likely not a simulator issue - but a circuit issue.

Shawn

Dear SpiceMonkey,

SpiceMonkey said:

Sometimes phase rises sharply and sometimes falls sharply, while their magnitude basicaly has no difference.Changing sweep points density, probe location or coner...all these may lead to this "controdictory" result.

Could anyone help explan this phonomeon? thank you!

Let me know if you think I am misreading your plot of magnitude and phase for the two cases you illustrate, but it certainly appears to me that there is a difference in magnitude that may lead to the two corresponding difference loop phase results. Please recall, first of all, that phase changes are far more sensitive to the frequency than magnitude changes. Hence, it is far more likely you find more significant phase changes than magnitude changes as frequency increases. Secondly, examining your magnitude plot around 10 MHz, there appears to be a significant resonance in the blue curve that is not as evident in the red curve. As a result, I might expect a much larger phase shift associated with the blue curve than with the red curve - and that is what you are observing due to the phaseWrapped() function used by the Calculator. Hence, in my opinion - and perhaps ny thoughts are incorrect in your opinion, your phase results are not totally unexpected to me and likely not a simulator issue - but a circuit issue.

Shawn

Thank you ShanwnLogan. Yeah phase is more sensitive than magnitude. While it's a litty tricky to analysis these zeros/poles which are beyond the GBW. Anyway basically no effect on transient response.

The loop gain is almost zero at the frequency where the phase "jumps"; if you do a polar plot of the (complex) loop gain, you will see that the absolute difference between the curves is very small.

Thank you Frank.