Re: [SI-LIST] : board-level simulation for differential signals

About this list Date view Thread view Subject view Author view

From: Jonathan Dowling ([email protected])
Date: Mon Apr 17 2000 - 17:11:19 PDT


Abe,
The easiest way to make arbitrary, coupled TLINE
geometries in XTK is to use the old *cfed* which
is still provided in the $QUADBIN directory.

CFED is an elaborate interface for XFX which allows
for graphical entry of geometry/material info. Try out
the conductor/dieletric rings! This tool will allow
you to explore the operating limits of XFX.

To invoke:

1. Make a working directory.
2. Copy $QUADHOME/viewlogic/prevue/defaults/defaults.xfx
    into the directory.
3. Type "cfed -d defaults.xfx"
4. Choose from default 1, 2, 3, 5 conductor configurations

It should handle configurations with up to 10 conductors
(or 5 diff pairs).

Jonathan Dowling

--- ARiazi <[email protected]> wrote:
> Hi:
>
> An interesting version of differential signal analysis using XTK involves
> appraisal of various types of stackup geometries, aimed at achieving
> desired range of target impedance values for Low Voltage Differential (LVDS)
> and Single Ended (SE) traces.
>
> A .xfx file, having different microstrip and stripline configurations, can
> be manually created. After execution of the XFX program a Transmission Line
> Parameters (.tlp) file results which contains values for the odd impedance
> (Zo) and even impedance (Ze). It is then possible to calculate impedance
> values for LVDS and SE signals from Zo and Ze, by means of simple formulas.
>
> Best Regards,
>
> Abe Riazi
>
> [email protected]
> [email protected]
>
>
> ----- Original Message -----
> From: Marc Humphreys <[email protected]>
> To: <[email protected]>
> Sent: Thursday, April 13, 2000 11:56 AM
> Subject: RE: [SI-LIST] : board-level simulation for differential signals
>
>
> > Daniel,
> >
> >
> > So as a follow-on, you had mentioned the use of XTK in your original
> post.
> > If
> > you are not familiar with all the capabilities XTK as a tool,
> > XTK will handle the coupling. As Weston mentioned it is an
> > important capability in simulating differential signals because
> > it speaks to the very reason one would use differential signalling - ie
> > noise immunity. Not only will XTK accurately calculate the single and
> > differential impedance on all the traces in the design but all the
> coupling
> > coefficients too - this is not a single ended vs. diffential problem,
> rather
> > in one
> > case you try to avoid it, in the other you exploit it.
> > When it comes to simulating coupling in the system (intentional or not) it
> > makes no difference to XTK what CAD system was used to route the etch
> > since XTK produces a CAD/Board neutral model
> > of the system interconnect for simulations purposes, from which it
> > determines, based
> > on proximity, what nets are coupled to what. No configuration is required
> to
> > indicate to XTK if a signal is coupled to neighboring nets,
> > wether it be a differential of single ended signal.
> > Noise couples between any two signals using the same principles,
> regardless
> > of intent,
> > and thats how XTK treats each and every net. Noise coupling and
> differential
> > coupling use the same physical principles.
> >
> > Additionaly (with a little creativity), simulating the coupling within
> > connectors
> > is not hard to do either with XTK.
> >
> > Marc
> >
>
>
>
>
>
> **** To unsubscribe from si-list or si-list-digest: send e-mail to [email protected].
> In the BODY of message put: UNSUBSCRIBE si-list or UNSUBSCRIBE si-list-digest, for more help,
> put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu
> ****
>
>

__________________________________________________
Do You Yahoo!?
Send online invitations with Yahoo! Invites.
http://invites.yahoo.com

**** To unsubscribe from si-list or si-list-digest: send e-mail to [email protected]. In the BODY of message put: UNSUBSCRIBE si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:36:13 PDT