The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> About initial transient in periodic steady state simulation
https://designers-guide.org/forum/YaBB.pl?num=1515142399

Message started by Jacki on Jan 5^th, 2018, 12:53am

Title: About initial transient in periodic steady state simulation
Post by Jacki on Jan 5^th, 2018, 12:53am

Hi,

Recently I am confused about my phase noise simulation of an oscillator. Because the oscillator has the reference bias circuit, the settling time is a bit long if all of the circuits is enabled at the beginning. Hence inside the ring oscillator, I add a delay cell to postpone the enable signal of the ring oscillator and let the reference bias circuit to settle at the beginning.
My question is: if I use periodic steady state simulation, is it possible to get the steady state in the simulation with the delayed enable signal for the oscillator.
If I can do it, I think I should take care the initial transient setup in the simulation. In Cadence, it is called tstab. I first run the transient simulation and make sure the oscillator can oscillate stably, then I set the initial transient time in the periodic steady state simulation. Am I doing in a good way?
From my view, the initial transient in the periodic steady state simulation will take care the delayed enable signal and the oscillator settling time. After the steady state is found, the phase noise simulation can start to run.
Please give me some hints. The reason I add the delayed enable signal for the oscillator is just to separate the settling between reference bias circuit and the oscillator.
Thank you very much.

Title: Re: About initial transient in periodic steady state simulation
Post by Ken Kundert on Jan 5^th, 2018, 9:48am

In SpectreRF, the PSS analysis naturally waits until the sources become periodic before starting the tstab interval. So the simulator should do what you want naturally with no special effort. This assumes that the sources are built-in source (not built with Verilog-A).

-Ken

Title: Re: About initial transient in periodic steady state simulation
Post by Jacki on Jan 21^st, 2018, 12:07am

Thank you very much, Ken. Currently I am not using SpectreRF unfortunately. I have solve the problem, and the simulation failure is due to the weakness of the software I am using. Thank you again for your clear explanation of PSS simulation.
Jacki

Title: Re: About initial transient in periodic steady state simulation
Post by Andrew Beckett on Jan 21^st, 2018, 9:10am

This is why it's always a good idea to explain which simulator you're using, which environment you're using, and the versions. In this case you just referred to "Cadence" which led Ken to (reasonably) believe that you were referring to Spectre (and the RF analyses of Spectre at that) within Virtuoso Analog Design Environment. Since there's no such tool as "Cadence" (this is the name of the company that provides the tool and not the name of the tool), it wasn't very clear what you were referring to...

Regards,

Andrew