[time-nuts] Spice simulation of PSRR and phase noise

Attila Kinali attila at kinali.ch
Fri Oct 27 14:19:48 EDT 2017


Hoi Dana,

On Sun, 22 Oct 2017 09:23:26 -0500
Dana Whitlow <k8yumdoober at gmail.com> wrote:

> But I am surprised about the simulation times that you speak of.
> Would you be willing to post some information detailing your
> methodology and an example "simple" circuit?

As Bruce already mentioned, the simulation times required for
getting decent results is 100-1000 times as long as the lowest
frequency considered. This is basically a statistical issue as
noise simulation, to be accurate, has to average over several
"runs" to remove effects of the noise source behaviour.

Another thing is, that, for the transient simulation to be accurate,
the maximum steps size has to be limited such that the maximum voltage
or current step seen is small. Ie if there is anything in the system
with high slew rates, then the step size has to be adapted to this
slew rate. This in turn makes it slower

Additionally, most spice implementations have quite short running
random number generators (usually with a state space of 2^16 to 2^32,
few with 2^64) which in turn requires some tricks to get decent results
out of it, that again make the simulation time longer.


				Attila Kinali

-- 
It is upon moral qualities that a society is ultimately founded. All 
the prosperity and technological sophistication in the world is of no 
use without that foundation.
                 -- Miss Matheson, The Diamond Age, Neil Stephenson


More information about the time-nuts mailing list