Re: [SI-LIST] : High Speed Queries

Fred Rosenberger (fred@gpc.ibc.wustl.edu)
Fri, 6 Nov 1998 16:56:40 -0600 (CST)

I strongly recommend the book on "Digital Systems Engineering" by
Dally and Poulton.

http://www.cup.org/books/dally/dally.html

Disclaimer: my financial compensation from this book was limited to a couple
of free copies.

Regards,

Fred

> From owner-si-list@silab.Eng.Sun.COM Fri Nov 6 14:47 CST 1998
> From: PRASAD_VENUGOPAL@HP-SanJose-om1.om.hp.com
> X-OpenMail-Hops: 1
> Date: Fri, 6 Nov 1998 12:23:10 -0800
> Subject: Re: [SI-LIST] : High Speed Queries
> MIME-Version: 1.0
> TO: portman@bitmicro.com
> CC: lweemei@dso.org.sg, si-list@silab.Eng.Sun.COM
> Content-Disposition: inline; filename="Re:"
> Content-Transfer-Encoding: 7bit
>
> Hi,
>
> I also recommend the folowing web site:
>
>
> http://www.ultracad.com/articles.htm
>
> ______________________________ Reply Separator _________________________________
> Subject: Re: [SI-LIST] : High Speed Queries
> Author: Non-HP-portman (portman@bitmicro.com) at HP-Boise,mimegw4
> Date: 11/6/98 10:20 AM
>
>
> Lum Wee Mei,
>
> Signal Integrity and PCB layout is not easy is different for each and every
> application, but it can be learned.
>
>
> The first thing to do purchase a copy of
>
> High-Speed Digital Design : A Handbook Of Black Magic
> By Johnson, Howard W. / Graham, Martin
> http://www1.clbooks.com/asp/bookinfo/bookinfo.asp?theisbn=0133957241
>
> CONTENT HIGHLIGHTS:
>
> Covers signal reflection, crosstalk, and noise problems that occur in
> high-speed digital machines (about 10 megahertz).
>
> Includes checklists that ask the questions an experienced designer would ask
> about a new system.
>
> Offers useful formulas for inductance, capacitance, resistance, rise time, and
> Q.
>
> Explains the trade-offs between signal crosstalk, mechanical fabrication of
> tolerances, and trace routing density.
>
> Presents a methodology for determining how many layers will be required to
> route a printed circuit board.
>
> ================================================================================
> ======
>
> The second thing to do is sit down for a couple undisturbed days and read it
> very carefully.
> Then read it again.
> ================================================================================
> ======
>
> Thirdly log onto Howard's web page www.sigcon.com - read, read, read.
> ================================================================================
> ======
>
> Fourthly, purchase a copy of the following book and read it thoroughly as
> well.
>
> EMC and the Printed Circuit Board: Design, Theory, and Layout Made Simple by
> Mark Montrose
>
> Contents : EMC Fundamentals; EMC Inside The PCB; Components
> and EMC; Image Planes; Bypassing and Decoupling;
> Transmission Lines; Signal Integrity and
> Crosstalk; Trace Termination; Grounding
> Copyright : 1999
> Format : Hardcover
> Pages : 320pp
> List Price US$ : $69.99
> IEEE Member Price US$: $56.00
> IEEE Product No. : PC5756-QAJ
> ISBN : 0-7803-4703-X
> http://www.ieee.org/products/ordinfo.html
> ================================================================================
> ======
>
> After all this reading, you can probably answer a lot of your own questions.
> But remember, there
> is not just one solution and it is not an easy subject matter. It takes
> knowledge, experience and
> some guesswork.
>
> Finally, may I suggest you get a Signal Integrity (SI) tool. There are
> several like Viewlogic's XTK
> and Mentor's IS. The one I like is Hyperlynx. (www.hyperlynx.com) Mainly
> because it is very easy to use,
> secondly because you don't have to sell your first born son to get it. It may
> not be as accurate
> as the previous two, but I don't have to hire a guru to run it.
>
> LineSim is a good first cut - but you have to lay out the signal topology
> yourself one at a time.
> It is ok for scratch pad work and pre-planning.
> BoardSim lays out all the signals at once by reading in your actual board
> layout from your CAD tool.
> BoardSim willl let you approximate things with the generic models it provides,
> but you WILL have to
> learn what IBIS models are before you can really make use of BoardSim.
> http://www.eia.org/eig/ibis/ibis.htm
>
> Download the demo and take a look.
>
> Like I said, SI and PCB layout is not easy, but with some reading and
> experience you can get along
> just fine.
>
>
> Good Luck,
>
>
> Roland
>
>
> Lum Wee Mei wrote:
> >
> > Hi, Todd and the rest in this mailing list:
> >
> > I am a "new kid in town" and is task to look into the high-speed design area.
> I understand this is not an easy topics and I hope to get some expert guidances
> from all of you. I have to come out with design rules so as that designs
> > require such area can refer to. Have someone draft a set of high speed design
> rules that can share with me?
> >
> > I have the following queries that that some of you might able to help me. They
> are:
> >
> > 1. How you identify high speed signal? By its rise time or by its operation?
> > 2. Someone told me that all high speed signals have to be impedance controlled
> . Is this correct? If not, what signals should be impedance controlled.
> > 3. I come across a so-called high-speed board which has many vias being punche
> d on its fill (for both ground and power) along side the pads for inductors and
> capacitors. They are also found on very thick ground traces. What are
> > the purposes of such vias? Did they play a role in high speed?
> >
> > I look forward to your reply.
> > Thanks and regards.
> >
> > **** 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-li
> st archives are accessible at http://www.qsl.net/wb6tpu/si-list ****
>
> --
> Roland F. Portman
> BiT Microsystems, Inc.
> 48499 Milmont Drive
> Fremont, CA 94538
>
> (510) 623-2341 Ext 129
> (510) 623-2342 Fax
>
> portman@bitmicro.com
> http://www.bitmicro.com
>
> **** 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 ****
>
>
> **** 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 ****
>
>

**** 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 ****