The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Non-Linear Stability Analysis and Load Lines
https://designers-guide.org/forum/YaBB.pl?num=1563837958

Message started by tec5c on Jul 22^nd, 2019, 4:25pm

Title: Non-Linear Stability Analysis and Load Lines
Post by tec5c on Jul 22^nd, 2019, 4:25pm

The non-linear stability analysis method through means of pole-zero identification states that the response is linearised about the large-signal periodic steady state (PSS) response. The literature also states that the HB simulation takes into account all the various states along the load line, when considering a power amplifier.

The pole-zero identification method makes use of a closed-loop transfer function. My question is, for every state along the load line there is (or should be) a corresponding transfer function. How can this method take various states along the load line into consideration and produce a single transfer function for stability analysis?

Title: Re: Non-Linear Stability Analysis and Load Lines
Post by Ken Kundert on Jul 23^rd, 2019, 2:42pm

If you are using SpectreRF and you are concerned about the stability of a feedback loop, then you would use the pstb analysis.

It is not computing large signal stability, but neither would the method you are describing. Instead, they report on the small-signal stability about a periodic operating point.

-Ken

Title: Re: Non-Linear Stability Analysis and Load Lines
Post by tec5c on Jul 24^th, 2019, 6:12am

Ken, thank you for your reply. However this doesn't answer my question. I'll try to simplify.

For a given PA, the device may be "taken" at any given point along the load line. How can a single transfer function of the circuit possibly "take all various load line states into consideration" as the literature claims?

Title: Re: Non-Linear Stability Analysis and Load Lines
Post by Ken Kundert on Jul 24^th, 2019, 11:19am

SpectreRF simply linearizes the circuit about the periodic operating point, and then performs the small signal analysis about that operating point.

Essentially normal stability analysis linearizes the circuit about a DC operating point before applying the small signals required by the analysis. SpectreRF linearizes about the periodic operating point. Thus, normal stability analysis determines whether a linear time-invarient approximation of the circuit is stable whereas SpectreRF determines whether the linear periodically varying approximation of the circuit is stable.

-Ken