[SI-LIST] Re: ibis limitations

Chris,

Who would beleive that we would ever agree?

Yes, Hspice does have issues with modeling loss, especially with lossy 
materials like FR4.  These can sometimes be resolved with w-element 
table models, but the root cause is generally non-causality caused by 
modeling dielectric loss and dielectric constant as fixed quantities.  

There is a great deal of research in the area of dielectric modeling at 
this time and much may come out of it.  But at this time, the best 
solution is use precise s-parameter measurements of the structures and a 
methodology to simulate them.  One is ADS, as you have mentioned. 
 Another is Nspice by Apache, which will support n-port s-parameters. 
 Another is Broadband Spice by Sigrity, which will convert n-port 
s-parameters into a Spice black box.   All will give exceedingly good 
results and are well beyond the capabilities of any IBIS board level 
simulator.

Best regards,

Scott


-- 
Scott McMorrow
Teraspeed Consulting Group LLC
2926 SE Yamhill St.
Portland, OR 97214
(503) 239-5536
http://www.teraspeed.com



Chris Cheng wrote:

>Well said Scott. And no one would probably image this come from Chris' mouth
>but even HSPICE seems to have significant problem modeling advanced
>packaging and long lossy interconnect with dielectric loss. A more advanced
>simulator like Agilent ADS seems to do the job better.
>Chris
>
> -----Original Message-----
>From:  Scott McMorrow [mailto:scott@xxxxxxxxxxxxx] 
>Sent:  Friday, July 26, 2002 4:05 PM
>To:    kalgudo@xxxxxxxxx
>Cc:    si-list@xxxxxxxxxxxxx
>Subject:       [SI-LIST] Re: ibis limitations
>
>
>Kal,
>
>An IBIS model will model a the driver characteristics of a 3 gig 
>differential pair driver, if that driver is modeled well by the model 
>supplier, if that driver does not have preemphesis, and if that model 
>does not have a package.
>
>IBIS cannot currently handle drivers with preemphesis and is very poor 
>at modeling packages. You will have significant divergence from reality 
>if you use an IBIS package model in a 3 gig differential simulation.
>
>The solutions is generally to use Spice to model your 3 Gig devices.
>
>Regards,
>
>Scott
>
>  
>


------------------------------------------------------------------
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 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: