AW: rising and falling time in IBIS and XTK

Jason,

You have to be aware, that in IBIS the rise and the falltime are only for the specific waveform.

When You look at the QUAD-Model, there is one risetime and one falltime for both rising/falling curves.

What happens is that if there is a delaytime for the lets say rising to GND waveform compared with the rising waveform to vcc, this delay will be added to the QUAD overall risetime.

In other words:
for IBIS

rise 50 gnd starts at 0ns and end at 5ns ( risetime will be 5ns )
rise 50 vcc starts at 0.5ns and ends at 6 ns ( risetime will be 5.5ns )

but the risetime in QUAD will now be 6ns

I hope this solves your problem

Eckhard

Eckhard Lenski
Siemens AG ICN M TC 6
Hofmannstr.51 81359 München
Tel : 0049 89 722 27776
Fax: 0049 89 722 44692
Email: eckhard.lenski@icn.siemens.de

> -----Ursprüngliche Nachricht-----
> Von: Jason Leung [SMTP:jleung@cid.alcatel.com]
> Gesendet am: Mittwoch, 7. März 2001 14:52
> Cc: ibis-users@eda.org
> Betreff: rising and falling time in IBIS and XTK
>
> Hello:
> I have a question regarding the rising and falling time, I understand
> that the ibis2xtk is going to translate the ibis model to Quad (xtk)
> model, but when I compare the original rising and falling waveform (ibis
> model) with the rising and falling waveform after the translation( Quad
> model), I have found a big difference in the rising and falling time
> (150 ps) between those two models.
> I know that the translator is going to remove the non-monotonic points
> inside the ibis model , in addition to that is there any more reasons
> why so large in the time difference.
> Is there any remedy to correct the rise time and make it closer to the
> values with the ibis model?
>
> I appreciate any feedback I can get in this regard.
>
> Thanks
>
> Jason Leung
> << Datei: Card for Jason Leung >>
 
Received on Wed Mar 7 06:48:52 2001