OPA277: Phase margin and stability

Adachi-san,

The information presented in the stability analysis section of the TI Precision Labs - Op amps sessions is based on classic control theory analysis techniques. Our TI op amps for the most part have a first order response to and beyond the unity-gain cross frequency (Aol 1 V/V). There may be a secondary pole beyond that point that can effect the phase margin. Then, it is the pole introduced by the capacitive load in conjunction with the op amp's complex output impedance (Zo) that introduces an additional pole creating a second-order system. Once that happens then the classic 2nd-order analysis techniques of Phase Margin vs Percentage Overshoot and Phase Margin vs AC peaking can be applied. The graphs come from textbooks on control systems theory.

The op amps and there loads may add additional poles and zeros to the ac response across frequency and that is where simulation models become a more accurate way to evaluate circuit stability. Through bench measurements and extensive simulations of the entire op amp circuit we are able to obtain accurate open-loop gain vs frequency (Aol) and complex output impedance (Zo). Once known, they can be designed into the simulation model and that with the increased mathematical capabilities of Spice an accurate ac response for the op amp in a particular circuit can be had.

"In the P-9, it says that the phase margin is 27degree when the AC Peaking is 6.7dB." When using the gain peaking to determine the phase margin of a second-order system you just apply the amplitude of the peak (6.7 dB) relative to the low frequency gain (0 dB) of this circuit. You don't use the phase portion of the simulation's gain-phase plot. It is the 6.7 dB gain peaking information that is entered into the Phase Margin vs. AC Peaking graph. That yields the 27 degree phase margin seen in the example.

Regards, Thomas

Precision Amplifiers Applications Engineering

Hello Thomas-san,

I understand. I appreciate it.

Regards,

Adachi