[SI-LIST] Fwd: Re: relation b/w impedance and SI
- From: karan bagga <kbagga31@xxxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Thu, 4 Sep 2003 23:50:18 -0700 (PDT)
Thanks Steve for the reply
I am some more doubts.....
1)
Is the impedance of the track only to do with matching it with the output
impedance
of driver or there are some other factors also while deciding impedance.
2)
Has impedance anything to do with the speed of the signal propagating it ?
Generally clocks are routed at 100ohms tracks and buses at 50-60 ohms ?
Is their any relation between impedance and speed ?
3)
while designing I/O impedance what are the considerations ?
Karan
Note: forwarded message attached.
---------------------------------
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
-- Attached file included as plaintext by Ecartis --
X-Apparently-To: kbagga31@xxxxxxxxx via 216.136.130.37; Thu, 04 Sep 2003
02:40:28 -0700
Return-Path: <weirsp@xxxxxxxxxx>
Received: from 204.127.202.55 (EHLO sccrmhc11.comcast.net) (204.127.202.55)
by mta122.mail.sc5.yahoo.com with SMTP; Thu, 04 Sep 2003 02:40:27 -0700
Received: from left.atdial.net
(12-236-230-135.client.attbi.com[12.236.230.135](untrusted sender))
by comcast.net (sccrmhc11) with SMTP
id <2003090409402701100gi6aae>; Thu, 4 Sep 2003 09:40:27 +0000
X-Sender: weirsp%atdial.net@xxxxxxxxxxxxxxx
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Thu, 04 Sep 2003 02:44:38 -0700
To: kbagga31@xxxxxxxxx
From: steve weir <weirsp@xxxxxxxxxx>
Subject: Re: [SI-LIST] relation b/w impedance and SI
In-Reply-To: <20030904091322.54056.qmail@xxxxxxxxxxxxxxxxxxxxxxx>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Content-Length: 1292
Karan,
Matching impedance helps both EMI, signal integrity, and in some cases
preservation of the components themselves. Grossly, excessive over /
undershoot can stress components to the point that they malfunction, or
even fail prematurely. BiCMOS memories are notoriously susceptible.
If you are designing with off-the-shelf parts, you are limited by what you
can get from those parts. So usually you will want to match your board
impedance to the driver. In fortunate circumstances, this results in less
circuitry and easier layout when matching eliminates the need for external
passive terminating components.
If you are designing with programmable ICs or the ICs themselves, you want
to be mindful of the application:
1. What range of impedances are practical for the likely board
layouts? This is usually contained between 35 and 55 ohms for common board
geometries and materials. Low K materials drive the impedance up. Thin
dielectrics drive it down.
2. What is the intended topology and loading? For a bus, the lower the
impedance you can tolerate, generally the better the signal integrity as
this swamps out the effects of parasitic package capacitance. Rambus for
example is down around 28 ohms.
For moderately fast edges, skin effect inserts series losses. For really
fast edges, dielectric present parallel losses. Going down in impedance
usually helps at the expense of more driver power.
3. What is the driver power dissipation? How much can it handle?
I hope this helps.
Regards,
Steve.
At 02:13 AM 9/4/2003 -0700, karan bagga wrote:
>Generally, In datasheets you will find the recommended Impedance for the
>BUS or a Signal.
>
>How do one decide upon the impedance ?
>
>Is it only related with matching the output impedance of Driver or it also
>helps reducing Signal Integrity Issues ?
>
>Regards
>Karan
>
>---------------------------------
>Do you Yahoo!?
>Yahoo! SiteBuilder - Free, easy-to-use web site design software
>
>------------------------------------------------------------------
>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:
>//www.freelists.org/webpage/si-list
>
>For help:
>si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>List archives are viewable at:
> //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
>
------------------------------------------------------------------
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:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//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:
