Re: [SI-LIST] : Stackup Extraction

Pat Zabinski (zabinski.patrick@mayo.edu)
Mon, 16 Aug 1999 12:21:38 -0500

Mitch,

You picked up on one key point: "design" vs "use".

If all you need is simple trace capacitance for your line analysis,
then Allegro (and other such tools) work fine. However, I have
found packaging parasitics, discontinutities, mutual coupling, etc.
to be a significant factor in SI, and these "parasitics" are not
included in any layout tool that I am aware of. (yes, Allegro
does have "SigNoise", this is another topic)

As such, I typically only "use" layout tools to implement what
I have already "designed" through simulation.

I'm sure there are several folks that co-layout and -design, but I
have found it easier to "design" first, then "layout" within
the design constraints.

Sorry for not being clear,
Pat

On Aug 16, 8:37am, Mitch Morey wrote:
> Subject: Re: [SI-LIST] : Stackup Extraction
>
> Pat Zabinski wrote:
>
> > #2 - I've been using Allegro for eight years for everything from
> > single chip packages to MCMs to large PCBs, and I have rarely
> > entered the material parameters into the tool. Although Allegro
> > does have an integral tool that spits out some basic parasitics
> > for lines (Z0, C, L, Td, etc.), I have never relied on it. The
> > only time I've ever entered in the material parameters/thicknesses
> > is for documentation purposes. Because it is not necessary
> > to enter in the cross section (i.e., it's an extra unnecessary
> > step), I'll venture a guess that many other designers avoid
> > entering it as well.
>
> Hi Pat,
>
> I really have to take issue to this! You say you've been using it for 8
years! Incredible! But you did say you've been "using" it, not DESIGNING with
it. If in your 8 years you have taken a little time to verify the results you
> would've found you can easily do more things with the results. Each time I
have an engineer come in a ask me what the capacitance on one of my traces, I
amaze them. Not that I have to sit and figure it out, but it's right there in
> front of me. I've done this on 2.4Ghz boards too. Want to know what my
impedance will be on a finished board? I can tell you, let me just select a
net! Does this take alot to set up? Well, with eight years of using the tool, I
> wouldn't know. But after I've set up my databases based on what my
fabricators do I can tell you what I'm designing to. Need resistivity of a
clock trace? I can tell you. etc, etc, etc.
>
> It's not hard! I'm not knocking you. I know many "designers" who are rowing
the same boat. But what I find bewildering is you have an educational
institution email address? I would think that out of most people, you certainly
> would've taken the time to find the answers.
>
> Happy "using".
>
> Mitch
>

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****