[SI-LIST] Re: : hspice field solver question

Stanley Chiu wrote:

> All, Thanks a lot!
> But maybe I did not detail my question here, sorry, ,,what I 'd like to
> know is that if I want to do a board level post simulation, I got all
> pcb stackup information already and the signals is running at 1.25gMhz
> on pcb and I need to get a rlgc table for the use of W element., I
> wonder..r
>  Is there any relationship between parameters in pulse function and
> frequency running on pcb??
>  Or whatever frequency running on pcb is noyhing to do with the pulse
> function??

Yes, and no.  There is a relationship between the Pulse function parameters
and the frequency being simulated.  What relationship this has to the actual
signals on your PCB, is up to you.

It is advisable to make the Pulse function include whatever frequencies you
wish to simulate.  I can't tell you what you want to simulate, but you might
want to use both high and low frequencies.  Generally speaking, your
simulations ought to include at least the fastest switching signals that
would appear on your board, and probably other cases too.

If you want to simulate with a periodic waveform and a frequency of 1.25GHz,
which has a period of 800ps, it might look something like this:

Vin in1 0 pulse ( 3.3v 0v 5ns 0.1ns 0.1ns 0.3ns 0.8ns )

which would be a "square" (actually trapezoidal) wave with a fundamental
frequency of 1.25GHz.

Or you might use a sine wave at 1.25GHz.

Either of these might be OK for strictly periodic signals, like clocks.  But
if it's a data signal, then:
 (a) if the data switches at a 1.25GHz rate, then for normal NRZ data its
fundamental frequency would be no more than half that, or 625MHz; and
 (b) you also ought to look at cases where the data changes less frequently.

If the "Vin" signal itself is what you have running around your board, you
should consider rounding its edges to make it more realistic.  Your real
signals won't have edges with sharp corners like SPICE's Pulse waveform
does.

A single pulse waveform contains a wide spectrum of signals that may go on
indefinitely (especially for SPICE's sharp-edged Pulse waveform).  Thus,
even your first example (Vin in1 0 pulse 3.3  0  5n  0.5n  0.5n  25ns) has
some content that extends well beyond 1.25GHz ... until it encounters either
an active device or a low-pass filter element.

The higher frequency content diminishes in amplitude depending on the
pulse's rise and fall times and pulsewidth.  In the theoretical limit where
the pulsewidth and rise and fall times approach zero, you would get the
ideal impulse function which includes all frequencies at equal amplitude
from DC to light and beyond.

I have never used Hspice's built-in field solver so I don't know if it has
any dependencies on other elements (such as the voltage source Vin).  One
might think that it should generate its RLGC table without regard to the
stimuli that you might apply to it, especially since the actual waveforms
that appear on the W-element might have a considerably different harmonic
content than the independent voltage sources do.  In other words, the Pulse
function and the RLGC table for the W-element *should* be independent of
one another (in my opinion, at least).

Regards,
Andy


------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List technical documents are available at:
                http://www.si-list.org

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: