Number of Harmonics in PSS analysis to determine the maximum time step

Dear RFGeek,

RFGeek said:

How can I control the maximum time step in the PSS analysis by adjusting the number of harmonics under 'Output harmonics' section in PSS analysis window for autonomous circuits?

There are up to two portions of a PSS analysis: an optional initial transient simulation ("tstab" portion) to allow the circuit response to achieve a solution close to its steady-state solution; and a second portion where the simulator attempts to converge on an estimate of the steady-state solution using the waveform at then end of the "tstab" portion as its starting point.

a. If you are interested in adjusting the time step of the initial "tstab" portion of the simulation and are using spectre or spectre APS, the paramter "maxstep" can be specified in one of the Transient analysis GUI options tab. If you are using Spectre X, it will not recognize the "maxstep" parameter as a means of adjusting its tilmestep. However, its tilmestep can be adjusted with the dynamic parameter steppreset as detailed in the On-line support troubleshooting article at URL:

https://support.cadence.com/apex/ArticleAttachmentPortal?id=a1O3w000009fo7PEAQ&pageName=ArticleContent&oMenu=People%20who%20viewed%20this%20also%20viewed

b. If, on the other hand, you want to adjust the tilmestep during the "pss" portion of the simulation where it attempts to converge on a periodic solution, either parameter "maxacfreq" on the PSS options tab "" (Figure 1) or adjusting the number of output harmonics on the PSS GUI (Figure 2) will accomplish this.

From the article at URL:

support.cadence.com/.../ArticleAttachmentPortal

Maxacfreq forces 5 timesteps in the period of the frequency set by maxacfreq (in pss options form), and the maximum harmonics sets maxstep to force 20 timepoints in the period of the highest harmonic of the pss (on the choosing analyses form).

From the above mentioned article if the PSS frequency is 1GHz and you set maxacfreq to 100GHz the maximum timestep will be defined as 5 timepoints in the period of 100GHz.
If you set the maximum number of harmonics to N, this forces 20 timepoints in the period of N*1 GHz.

In this case, the timestep will be the same for both methods with a maxacfreq of 100 GHz or setting the maximum number of harmonics to 25. In essence the maximum timestep in both cases is:

(1/5)*(1/100 GHz) = (1/20)*(1/(25*1 GHz) = (1/20)*(1/25 GHz) = 2 ps

Does this help RFGeek?

Figure 1

Figure 2